MLinfo | 機械学習・AI論文まとめ

用途: ゲームの支援ツール
難易度: Easy
コスト: Medium

githubGitHubあり2026-11-04

ART — Agent Reinforcement Trainer: train multi-step agents for real-world tasks using GRPO. Give your agents on-the-job training. Reinforcement learning for Qwen3.6, GPT-OSS, Llama, and more!

ARTは、多段強化学習トレーナーです。このトレーナーは、GRPOを使用して、現実世界のタスクに対して、多段強化学習を行うことができます。

用途: 多段強化学習トレーナー
難易度: Easy
コスト: High

Reciprocity Can Erase Direction:tructural Boundaries and Controlled Utility of Magnetic Graph Networks

Magnetic graph networks encode direction through phase, but a learned graph can erase the intended directional

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

自然言語処理ファインチューニング回帰テキスト強化学習

arxivGitHubあり2026-11-03

Where Did the Update Go? Certifying Declared Change Envelopes for Model Updates

Deployed language models are updated continually, and regulators increasingly accept changes that were declare

用途: 回帰
難易度: Hard
コスト: High

ModelBlaster: Closing the Performance Gap through Learning-Augmented Agents Across Models, Kernels and Hardware Backends

Optimizing accelerator code across hardware architectures is difficult. ML compilers search predefined spaces

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

表形式向き説明可能CPUで試しやすい強化学習方策勾配 (PPO / A3C)分類画像テキスト

Statistical Machine Learning for Medicinal Plant Leaves Classification

Medicinal plants are usually identified using practitioners' experience, but reproducible, image based methods

用途: 分類
難易度: Hard
コスト: High

Harness-S: Learning from Sibling Harnesses for Generalizable Agent Training

Harness evolution from a common base exhibits Locality, Complementarity, and Nonstationarity: variants repair

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

Neologism Inoculation Prompting

Inoculation prompting is a method to reduce undesired behaviours such as reward hacking in frontier models. It

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

Shared Context Candidate Scoring with Jointly Adapted Readout Networks

Repeated decisions over a shared document create two distinct opportunities: reusing document computation and

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

Online iLQR-MPC with Whole-Body Control for Robust Humanoid Locomotion

Humanoid locomotion requires deciding where to step and how to distribute contact forces. However, a disturban

説明可能強化学習モデルベース

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

Grasping and Stabilizing Tumbling Objects with Rotating-Base Multi-Arm Space Manipulators

This paper presents an innovative guidance and control method for a space manipulator that actively maneuvers

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

VARP: A Visual Adaptive Rearrangement Planner for Non-Monotone Object Rearrangement Planning

Object rearrangement planning in cluttered environments is often non-monotone, requiring objects to be tempora

強化学習画像

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

githubGitHubあり2026-11-03

gradio — Build and share delightful machine learning apps, all in Python. 🌟 Star to support our work!

Pythonでマシンラーニングアプリを作成・共有することができるライブラリです。

用途: マシンラーニングアプリ作成
難易度: Easy
コスト: Medium

githubGitHubあり2026-11-03

machine-learning-for-trading — Code for Machine Learning for Trading, 3rd edition — from data sourcing to live execution.

LLMの推論 Transparency を高めるために、DiffusionGemmaの計算を分離しVariable Transparency とAlgorithmic Transparencyを評価します。

用途: LLMの透明性、誤用、過度安定化を理解する
難易度: Easy
コスト: High

AlgoPilot: Cross-Paradigm Reasoning in Language Models via Strategy Selection and Guidance

Real-world problems span diverse domains, including program synthesis, symbolic reasoning, planning, and optim

自然言語処理大規模言語モデル生成テキスト強化学習

用途: 生成
難易度: Hard
コスト: High

Finite-Sample Convergence in Networked Average Reward MARL: Decentralization Pitfalls and Entropy Remedies

We study the problem of average reward Multi-Agent Reinforcement Learning (MARL) where agents interact within

自然言語処理RAG強化学習

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Low

少数データ向きCPUで試しやすい条件最適化コンピュータビジョン物体検出強化学習

Forgetting to Improve: Principled Data Removal in Active Learning

The uncertainty of a statistical model is most commonly factorized into an aleatoric and an epistemic part. Th

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

品質予測/異常検知深層学習Transformer生成テキスト強化学習

Language Model Augmented Decision Transformer for Multi-Task Offline Reinforcement Learning

Offline multi-task reinforcement learning (MTRL) aims to learn a unified policy from heterogeneous offline dat

用途: 生成
難易度: Hard
コスト: High

少数データ向き深層学習Transformer生成強化学習

Decision Transformer with Task-Aware Representations for Offline Meta-Reinforcement Learning

To enable effective generalization in offline meta-RL, it is essential to accurately infer the underlying task

用途: 生成
難易度: Hard
コスト: High

EpiLoRA: Fusing Many LoRA Adapters into Low-Precision GEMMs for Serving and RL

Low-rank adapters (LoRA) let one base model serve many fine-tuned variants: multi-tenant engines keep a single

自然言語処理大規模言語モデル

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

センサ/時系列品質予測/異常検知強化学習方策勾配 (PPO / A3C)検出回帰異常検知

LaplaceDiag: Fault Diagnosis through Changes in Learned Actuator Effects

Identifying faulty components from sensor anomalies is difficult because fault effects can propagate through a

用途: 検出
難易度: Hard
コスト: High

AIRoA-0: Long-Horizon, Precise, Cross-Embodiment Manipulation from Handheld Data

AIRoA-0 (A-0) is a bimanual manipulation system learned from 16,151 hours of handheld demonstrations, with no

自然言語処理RAG動画マルチモーダル強化学習

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

WebHarbor: Docking Real Websites for Evolving GUI Agent Environments

GUI web agents require realistic and interactive environments that remain stable during both evaluation and tr

深層学習軽量化・量子化画像強化学習

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

自然言語処理大規模言語モデル異常検知テキスト強化学習

Not All Reasoning Generalizes: A Case Study on Learning Generalizable Reasoning Strategies in RLVR

Reinforcement learning with verifiable rewards (RLVR) can enable generalizable improvement for large language

用途: 異常検知
難易度: Hard
コスト: High

FADE: Factored Ability--Difficulty Filtering for Online RLVR Scheduling

Group-based reinforcement learning with verifiable rewards (RLVR) learns only from prompts whose sampled compl

自然言語処理RAG生成強化学習

用途: 生成
難易度: Hard
コスト: Low

D&A: Decoupling Description and Assessment Enables VLMs as Zero-Shot Robotic Reward Models

Vision-Language Models (VLMs) hold promise for zero-shot video task-progress estimation in robot learning, yet

深層学習軽量化・量子化検出テキスト動画

用途: 検出
難易度: Hard
コスト: High

Spacetime Emergence in the Memory of a Learning Agent

We show that the causal structure of spacetime can emerge within the memory of a learning agent performing a q

強化学習モデルベース

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Bridging Neural Tangent Kernel Analysis and Reinforcement Learning for Deep Function Approximation

Despite their widespread success across machine learning, the learning dynamics and generalization properties

深層学習軽量化・量子化教師あり強化学習

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

An Axiomatic Framework for Reinforcement Learning in Open Multi-Agent Systems

Open multi-agent systems, where only a subset of agents are controllable, are central to real-world domains su

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Inferring Transferable Rewards via Active Inverse Reinforcement Learning

Inverse Reinforcement Learning (IRL) aims to recover expert rewards that transfer across environments with dif

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

DeepWaveRL: Self-Supervised Full Waveform Inversion via Reinforcement Learning

Full Waveform Inversion (FWI) is a fundamental inverse problem in geophysics to estimate subsurface properties

自然言語処理RAG教師あり自己教師強化学習

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

The Cost of Mismatch: Noise Amplification in Zeroth-Order Reinforcement Learning

Zeroth-order (ZO) optimization has emerged as a promising approach for memory-efficient large language model (

深層学習軽量化・量子化テキスト強化学習

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

From PID to Reinforcement Learning in Building HVAC Control: A Control-Architecture Review of Tuning Strategies, Formulations, and Deployment Trade-offs

This paper reviews control architectures for building heating, ventilation and air-conditioning (HVAC) systems

センサ/時系列強化学習モデルベース生成

用途: 生成
難易度: Hard
コスト: Medium

Controlling Sharpening in LLM Post-Training

LLM post-training via reinforcement learning has shown consistent improvements in accuracy at the cost of redu

自然言語処理大規模言語モデル生成強化学習

用途: 生成
難易度: Hard
コスト: High

品質予測/異常検知強化学習方策勾配 (PPO / A3C)生成

Optimal Subgroup Discovery at Every Support Threshold

Subgroup discovery aims to identify regions of the feature space where a target variable deviates from its mar

用途: 生成
難易度: Hard
コスト: Medium

Bayesian Sequential Optimal Experimental Design for Single-Pixel Imaging using Reinforcement Learning

Adaptive compressed sensing (CS) aims to select the most informative measurements sequentially, which is cruci

センサ/時系列深層学習軽量化・量子化画像強化学習

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

表形式向き機械学習表形式データテキスト表形式強化学習

Improved regret bounds in tabular RL under local differential privacy

We study regret minimization for time-homogeneous episodic tabular reinforcement learning with $S$ states and

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Low

強化学習方策勾配 (PPO / A3C)テキスト音声マルチモーダル

Unimodal Augmentation-based SSL Objectives Induce Cross-Modal Alignment

Recent works have observed implicit cross-modal alignment between independently trained unimodal models. Howev

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

Gibbs Variational Last Layers for Long-Horizon Dynamics Learning

Accurate long-horizon probabilistic predictions of dynamical systems are a prerequisite for model-based decisi

品質予測/異常検知強化学習モデルベース

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

A Universal Policy Wrapper With Guarantees

Many control tasks require driving the system into a prescribed goal set and subsequently keeping it there. Th

深層学習Transformer

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

On Stochastic Continuous-Time Linear System Identification for Adaptive Stabilization

Learning to stabilize an unknown continuous-time stochastic linear system is a central problem in control theo

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

On the Primacy Bias in RLVR Training

Reinforcement Learning with Verifiable Reward (RLVR) has emerged as a promising approach for enhancing the rea

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

品質予測/異常検知強化学習方策勾配 (PPO / A3C)生成テキスト

Characterizing Adversarial GEO via Implicit Brand Injection

Generative search engines increasingly shape how users discover products, services, and commercial entities by

用途: 生成
難易度: Hard
コスト: Low

Four Bridges: Measuring Deception Propensity in a Multi-Agent Game

AI agents are increasingly deployed in settings where they interact and coordinate with other agents, yet hone

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

An Agentic Framework for Automated Power System Studies and Grid Connection Assessment

Connecting new generation and demand to the grid depends on two scarce skills: checking applications against a

自然言語処理RAG生成

用途: 生成
難易度: Hard
コスト: Low

SQLoom: Set-Contrastive Learning of Syntactic, Semantic, and Logical Dependencies for Text-to-SQL

Text-to-SQL translates natural language questions into SQL queries, enabling database access without writing S

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

Axion 1.0: Open Agent Foundation Models, Simplified

We introduce Axion 1.0, a fully open agent foundation model that combines strong and broad agentic capabilitie

深層学習軽量化・量子化生成強化学習

用途: 生成
難易度: Hard
コスト: High

MIRA: Reinforcing Multimodal Reasoning via Deceptive Contextual Augmentation

While Reinforcement Learning with Verifiable Rewards (RLVR) has significantly advanced multimodal reasoning, e

自然言語処理RAG画像テキストマルチモーダル

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

BLPO: Beyond-Local Policy Optimization for Long-Horizon LLM Agents

Long-horizon tasks require large language model (LLM) agents to coordinate interdependent actions, a capabilit

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

Offline Approximate Value Iteration: A Reinforcement Learning Approach for Satellite Constellation Deployment

This paper introduces a reinforcement learning-based approximate dynamic programming approach to optimize the

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

品質予測/異常検知深層学習軽量化・量子化生成画像テキスト

OmniOPD: Stable On-Policy Distillation for Joint Audio-Video Generation

Reinforcement learning (RL) elevates joint audio-video generation, but optimizing heterogeneous, modality-spec

用途: 生成
難易度: Hard
コスト: High

Analysing the Narrative Paradox through Formal Models and Properties in Interactive Narrative

Interactive narratives aim to let players influence the unfolding of a story while preserving a coherent and s

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Dory1.0-4B: A Looped Normalized-GPT

We introduce Dory1.0-4B, a 4-billion-parameter looped (latent-recurrent) language model with elastic depth: a

深層学習Transformerテキスト強化学習

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

Dhārā Bench: An Evidence-Oriented Retrieval Benchmark for Indian Knowledge Systems

Dhārā Bench is a retrieval benchmark for Indian Knowledge Systems (IKS), with 325 questions in 12 categories a

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Low

Meet Students Where They Are: Student-Prefix Teacher Adaptation for On-Policy Distillation

On-policy distillation (OPD) mitigates the student's exposure bias by optimizing on student-generated trajecto

深層学習軽量化・量子化テキスト強化学習

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Low

A State Change Is Local, Its Consequences Are Not: Dependency-Aware State Revision for Long-Term Agent Memory

Long-lived agents rely on persistent memory of users, tasks and decisions, and this memory must stay consisten

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

品質予測/異常検知深層学習TransformerQAテキスト教師あり

EffComp: Efficient Prompt Compression via Hybrid Reinforcement & Supervised Learning

Prompt compression is aimed at reducing input prompt lengths to enable cheaper, faster, and potentially higher

用途: QA
難易度: Hard
コスト: High

EgoActionBench: Benchmarking Dense Action Captioning for Egocentric Human Videos

Learning from human data offers a scalable path toward embodied intelligence, yet it requires models to accura

自然言語処理ファインチューニング動画強化学習

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

強化学習方策勾配 (PPO / A3C)画像テキスト音声

TaoMate-FVSR:Audio-Guided Facial Video Restoration

AI-generated videos often depict people within detailed scenes while leaving their small facial regions poorly

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

品質予測/異常検知コンピュータビジョンセグメンテーション生成画像テキスト

Do Visual RL Gains Scale at Test Time? A Quantile View of Post-Training

Reinforcement learning (RL) post-training improves the single-sample quality of visual generative models, but

用途: 生成
難易度: Hard
コスト: High

Physical Experience Builds Continual Spatial World Models

World models are increasingly used to support prediction and planning in embodied agents, but the knowledge is

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

When Robots Outweigh Objects: How Inertia Mismatch Shapes Robot Policy Learning

Conventional robot arms can be extremely heavy compared with the everyday objects they manipulate, creating a

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

SNAP: Semantic Navigation Among Pedestrians

Navigating crowded urban environments requires robots to avoid pedestrians while adhering to social norms such

自然言語処理プロンプトエンジニアリング強化学習

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

センサ/時系列強化学習方策勾配 (PPO / A3C)分類音声

OpenZilo: An Open Smart-Ring Platform for Human-Centric Sensing and Interaction with Robots

Wearable robot interfaces must connect user input to timely physical action. We present the proposed platform,

用途: 分類
難易度: Hard
コスト: Low

Safe-Ballbot: Tilt-Consistent MPC with Control Barrier Functions for Ballbot Navigation

Ball-balancing robots (ballbots) must lean to accelerate, and the lean is limited by the wheel torques and by

強化学習動画

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

TRAILS: Trajectory Generation for Automated Indoor Locomotion Studies

Virtual reality experiments involving navigation and simulated robot motion often require large, realistic env

強化学習生成

用途: 生成
難易度: Hard
コスト: Medium

FineCraft: Enhancing VLA Models for Fine-Grained Manipulation

Fine-grained manipulation couples instruction grounding, local visual precision, and sustained execution. We p

強化学習画像

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

自然言語処理プロンプトエンジニアリングテキスト強化学習

KlaskTron: An Open-Source Platform for Physical Adversarial Multi-Agent RL

Progress in robot learning, particularly physical multi-agent reinforcement learning (MARL), is currently chal

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

少数データ向きセンサ/時系列強化学習方策勾配 (PPO / A3C)検出画像3D

A Tightly Coupled Magnetic-Inertial SLAM System for Field Robots

Field robots operating in mines, tunnels, and disaster areas require reliable localization despite GNSS denial

用途: 検出
難易度: Hard
コスト: High

説明可能自然言語処理プロンプトエンジニアリング強化学習

KlaskTron-IL: A Contact-Rich Adversarial Imitation Learning Benchmark with a Validated Sim-to-Real Bridge

KlaskTron is an open-hardware physical platform for two-player adversarial reinforcement learning, with a lear

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

BeyondContact: A Reinforcement Learning Framework for Airflow-Based Non-Contact Manipulation

Contact-based robotic manipulation is constrained by end-effector reachability and the mechanical properties o

MI向き深層学習軽量化・量子化強化学習

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

ReContact: Reconstruct and Retarget Human Contacts from Video for Contact-Rich Humanoid Manipulation

Contact transitions allow humanoid robots to redistribute object support across the whole body, enabling manip

強化学習方策勾配 (PPO / A3C)動画

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

Positive Safety Margins from Artificial Potential Field Reward Augmentation for Size-Robust Obstacle Avoidance under Partial Observability

Collision-free reaching is central to manipulation, yet real cells often leave the size and exact position of

強化学習異常検知

用途: 異常検知
難易度: Hard
コスト: High

RESONATE: a Real Social Navigation Testbed for Autonomous Service Robots

Social navigation is essential for autonomous service robots operating in human environments, but progress rem

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

An intelligent satellite must do nothing well

Autonomous inspection of a resident space object is usually posed as a contest between classical next-best-vie

品質予測/異常検知強化学習モデルベース

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Beyond Policy Patches: Distilling RL Specialists into Vision-Language-Action Policies

Vision-language-action (VLA) policies can handle noncritical stages of long-horizon manipulation yet fail duri

深層学習軽量化・量子化マルチモーダル強化学習

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

Closing the Dynamics Gap through Self-Directed Robotic Interaction

Agentic Real2Sim2Real systems can reconstruct a scene in simulation from camera observations, but contact-rich

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

MI向き強化学習方策勾配 (PPO / A3C)画像

DexArena: A Unified Benchmark for Long-Horizon Dexterous Manipulation

Long-horizon dexterous manipulation remains a fundamental challenge for robot learning, requiring the composit

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Sequential Recovery Updates for Learned Controllers under Successful-Trajectory Preservation

A competent learned controller already recovers from many disturbances. Improving it means learning a correcti

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Touching Projected Agents through the DAT Screen -System Design and Questions of Interaction Boundaries-

Touching a virtual agent involves more than detecting a finger on a display: the contact surface can deform an

センサ/時系列強化学習

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

A Study on the Effects of Finger Movements on a Humanoid Hugging Robot

This study aimed to examine the impact of finger movements in hugging robots on users by experimentally invest

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

To Coordinate or Not to Coordinate? Learning to Swim with Eight Soft Arms

Does effective swimming with multiple underactuated soft arms require explicit coordination between arms? We h

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

少数データ向きCPUで試しやすい品質予測/異常検知条件最適化自然言語処理プロンプトエンジニアリング強化学習

Learning Purposeful Co-designs of Bio-inspired Quadrupedal Robots

Treating morphology and control as separate design problems has constrained the efficiency and adaptability of

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

At What Cost? (In)efficiency in First-Price Procurement Auctions

We study the efficiency of Bayes-Nash equilibria in first-price procurement auctions. We measure efficiency by

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Classification Aggregation with Multiple Memberships

People face the problem of classifying objects into categories while aggregating their opinions. We axiomatica

強化学習方策勾配 (PPO / A3C)分類

用途: 分類
難易度: Hard
コスト: Low

Coalitions, Stability, and Majority Jackpots: Frameworks for Unifying Majoritarian and Proportional Allocation in Multiwinner Elections

Social choice literature on multiwinner elections tends to focus on characterizing two distinct tasks: majorit

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

githubGitHubあり2026-11-02

annotated_deep_learning_paper_implementations — 🧑‍🏫 60+ Implementations/tutorials of deep learning papers with side-by-side notes 📝; including transformers (original, xl, switch, feedback, vit, ...), optimizers (adam, adabelief, sophia, ...), gans(cyclegan, stylegan2, ...), 🎮 reinforcement learning (ppo, dqn), capsnet, distillation, ... 🧠

🧑‍🏫 60+ Implementations/tutorials of deep learning papers with side-by-side notes 📝; including transformers

深層学習Transformer強化学習

用途: 実装・検証基盤
難易度: Easy
コスト: Medium

品質予測/異常検知強化学習方策勾配 (PPO / A3C)

Reduced Cost Influence Functions for Predict-then-Optimize under Noisy Data

Machine learning models deployed in decision-making pipelines, from vehicle routing to dynamic pricing to clin

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

Rare Event Constrained Reinforcement Learning with Large Deviations

We study finite-horizon reinforcement learning with a safety constraint on the probability of reaching a failu

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Direct Knowledge Transfer for Offline Q-Learning

Data-driven sequential decision making in marketing and healthcare can benefit from observations collected in

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

Near-Optimal $\sqrt{K}$-Regret for Online Learning in Uninformed Markov Games

We study online learning in two-player uninformed Markov games, where the opponent may follow an arbitrary, un

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

説明可能強化学習方策勾配 (PPO / A3C)分類

Surrogate Losses for Fairness-Aware Classification: Fisher Consistency and Calibration

Fairness-aware classification seeks to balance predictive accuracy and fairness across demographic groups. Alt

用途: 分類
難易度: Hard
コスト: Low

Retrieval-Enhanced Cultural Evaluation for Social Media Translation Training

Social media translation requires preserving meaning while capturing the cultural nuances of user-generated co

自然言語処理大規模言語モデル分類強化学習

用途: 分類
難易度: Hard
コスト: High

Don't Guess, Ask Wisely: Benchmarking and Training Proactive Elicitation in Open-Domain LLM Conversations

When user queries are incomplete, large language models should ask targeted questions before committing to an

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

深層学習Transformer画像テキストマルチモーダル

NAS-PO: Native Attention-Strategy Policy Optimization for Vision-Language Models

Reinforcement learning with verifiable rewards (RLVR) has become an effective post-training paradigm for visio

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

When to Probe and When to Act: Clarity-Aware Risk-Adaptive Control for Interactive Driving

Safe navigation in interactive traffic requires autonomous vehicles to reason about unobservable human charact

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

ODE Policy: Robot Learning via Action Dynamics, Not Actions

Most robot policies, whether trained by imitation learning (IL) or reinforcement learning (RL), map observatio

深層学習軽量化・量子化マルチモーダル強化学習

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

Beyond Validation Loss: Coverage and Reliability Dynamics in Robot Policy Pretraining

In modern robot learning, generative robot policies are first pretrained through supervised learning from demo

自然言語処理RAG生成画像教師あり

用途: 生成
難易度: Hard
コスト: High

Dual Path Following for an Aerial Manipulator Under State and Input Constraints via Transverse Feedback Linearization

This paper studies dual path following for a quadrotor carrying a three-degree-of-freedom arm. The center of m

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Adaptive Surface Cleaning under Uncertain Cleaning Response

Surface cleaning requires robots to adapt both where they clean and how they interact with contamination whose

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Optimal Online Learning in Bilateral Trade Through Interest

We consider online learning in bilateral trade on platforms such as Uber, Airbnb, and eBay. In each round, a p

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

The Stability of Data Exchange in Competitive Markets

We study the stability of data exchange among competing firms. Unlike tangible goods, data is non-rival: shari

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

huggingfaceHugging Faceあり2026-11-01

NorthCUA Technical Report: Turbocharging Computer-Use Agents through Experience at Scale

Scaling computer-use agents from individual actions to complex software work requires experience that connects

品質予測/異常検知自然言語処理ファインチューニング生成強化学習

用途: 生成
難易度: Easy
コスト: High

少数データ向き条件最適化コンピュータビジョンセグメンテーション分類テキスト強化学習

HALO: Hierarchical Evidence Selection and Reinforcement Learning for Allocating Budgeted Label Optimization from Longitudinal Clinical Records

Learning from extensive longitudinal clinical records--spanning repeated encounters, evolving conditions, and

用途: 分類
難易度: Hard
コスト: Low

Optimal Learning in Fairness-Aware Contextual Bandits

Mitigating unfair disparities in sequential decision-making is a growing concern in machine learning. We study

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Where Causal Discovery Breaks on Real Data: Measuring the Assumptions One Property at a Time

Causal discovery is usually evaluated on simulators or on systems whose ground truth follows from physics or d

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Differentiable Iterative Learning Control for Black-Box Position-Commanded Manipulators

High-accuracy repetitive tracking is challenging for position-commanded manipulators whose low-level controlle

強化学習モデルベース

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Hi-CoLA: High Confidence Lower Bound Approximation Based Reinforcement Learning for Flex-Route Transit Operation Control

This paper studies safe policy improvement for real-time operation control of flex-route transit (FRT) systems

説明可能深層学習軽量化・量子化強化学習

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Low

Generalizable robotic manipulation through compositional skill reuse

Hao Ren, Shuohan Chen, Senyuan Mu, Ruihan Huang, Wanfeng Shang, Yajing Shen

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Why Koopman Structure Does Not Help Differentiable World Models for Humanoid Whole-Body Tracking

We test whether Koopman structure, linear dynamics in a lifted state space, helps when a humanoid whole-body t

強化学習モデルベース

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

Trading Transparency for Stability: Friction-Cone-Constrained Bilateral Teleoperation for Opening Doors of Unsecured Furniture

Opening the door of unsecured furniture loads the frictional contacts at its supports; too large a pull slides

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Roz: A Taste Test for World Model Latents

State of the art latent dynamics world models achieve impressive performance for robot planning. But do they a

強化学習モデルベース画像動画

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

自然言語処理ファインチューニングマルチモーダル強化学習

Behavioral Steering of Vision-Language-Action Models in the Low-Parameter Regime

Reinforcement learning (RL) is increasingly used to adapt pretrained vision-language-action models (VLAs). We

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

How Much Can Approximate Equilibria Change the Social Cost of Selfish Routing?

We study approximate Nash equilibria in nonatomic congestion games, with the goal of understanding how much th

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Transformer-Based Reinforcement Learning for Phase I Dose Allocation

Phase I dose-finding trials must learn from sparse toxicity observations while limiting exposure to overly tox

深層学習Transformer回帰強化学習

用途: 回帰
難易度: Hard
コスト: Medium

表形式向き深層学習Transformer生成表形式強化学習

Statistical Foundations of Reinforcement Learning: A Non-Asymptotic Perspective

Reinforcement learning (RL) embodies a foundational paradigm for sequential decision-making in the face of unc

用途: 生成
難易度: Hard
コスト: Low

Local Budget Benefits, Population Commitment, and Solver Performance: A DGO Case Study with Controlled DE Transfer

An evaluated candidate can improve a local search step, alter the population used by later steps, and affect a

品質予測/異常検知強化学習マルチエージェント

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

When appearance and motion disagree: evaluating class--motion consistency against disguised uncrewed surface vessels

Defense systems for maritime assets must distinguish approaching benign craft from uncrewed surface vessels (U

強化学習方策勾配 (PPO / A3C)分類検出

用途: 分類
難易度: Hard
コスト: High

ROOF: Learning Humanoid Fall Recovery under Overhead Constraints

A fallen humanoid beneath an overhead obstacle faces a coupled recovery problem. It must create clearance whil

生成AIGAN強化学習

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

少数データ向きコンピュータビジョンセグメンテーション強化学習

Online Learning of Heterogeneous Locomotion Dynamics under Reinforcement Learning Policies for Precise Loco-Manipulation

How to reuse a pretrained locomotion policy for precise loco-manipulation is still an open problem. We propose

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

Integrating Low-Thrust Transfer Dynamics into POMDP Planning for On-Orbit Servicing

Recent developments in both high-thrust (HT) and low-thrust (LT) propulsion technologies broaden design option

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

arxivGitHubあり2026-10-30

OpenVIO: A Versatile Open-Source Visual-Inertial Odometry System

Visual-inertial odometry (VIO) is a fundamental topic in robotics research. Beyond research, VIO has been adop

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

ResMerge: Residual Policy Learning and Merging for Continual Adaptation of Pre-trained Robot Policies

Continual adaptation to new tasks is essential for robots deployed in the real world, but adapting pre-trained

説明可能深層学習軽量化・量子化強化学習

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

When Is a Game a Markov Decision Process? Correspondences Between Game Theory and Reinforcement Learning

Reinforcement learning and game theory describe the same object from two sides. Game theory says what a ration

自然言語処理大規模言語モデル音声強化学習

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

The Competitive Ratio of Online Budgeted Allocation is Strictly Smaller than $1-1/e$

A well-known conjecture in online algorithms is that the (randomized) competitive ratio of online budgeted all

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

huggingfaceHugging Faceあり2026-10-30

Latent Reasoning in Continuous Space for Unified Multimodal Models

Unified Multimodal Models (UMMs) achieve remarkable success on diverse multimodal understanding and generation

品質予測/異常検知自然言語処理大規模言語モデル生成画像テキスト

用途: 生成
難易度: Easy
コスト: High

githubGitHubあり2026-10-30

qlib — Qlib is an AI-oriented Quant investment platform that aims to use AI tech to empower Quant Research, from exploring ideas to implementing productions. Qlib supports diverse ML modeling paradigms, including supervised learning, market dynamics modeling, and RL, and is now equipped with https://github.com/microsoft/RD-Agent to automate R&D process.

クエンティング投資プラットフォームを実現するためにAI技術を活用します。

強化学習方策勾配 (PPO / A3C)教師あり

用途: クエンティング投資プラットフォーム
難易度: Easy
コスト: Medium

githubGitHubあり2026-10-30

ml-agents — The Unity Machine Learning Agents Toolkit (ML-Agents) is an open-source project that enables games and simulations to serve as environments for training intelligent agents using deep reinforcement learning and imitation learning.

Unityを使用してマシンラーニングエージェントを訓練して訓練できるツールです。

コンピュータビジョン3D・点群3D強化学習

用途: Unityでマシンラーニングエージェント
難易度: Easy
コスト: High

Density-Guided Bootstrap Boundaries for Latent World-Model Reinforcement Learning

Latent world models allow reinforcement learning agents to learn policies from imagined trajectories, yet dete

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

Reinforcement Learning Under Runtime Confounding: Policy Learning and Off-Policy Evaluation

Reinforcement learning (RL) is increasingly used to train robust, adaptable systems across healthcare, e-comme

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Deep Reinforcement Learning for Blind Passive Beamforming in RIS-Assisted MIMO Systems

Reconfigurable Intelligent Surfaces (RIS) are a promising solution for improving millimeter-wave (mmWave) comm

強化学習モデルフリー (DQN / SAC)音声

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Safe Multi-Objective Reinforcement Learning: Zero Constraint Violation and Primal-Averaging Convergence

Many applications of multi-objective reinforcement learning (MORL) require policies that balance conflicting o

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Low

Satisficing Regret Minimization in Bandits with Knapsacks

We study satisficing regret minimization in bandits with knapsacks (BwK), where a learner sequentially selects

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

コンピュータビジョンセグメンテーションテキスト強化学習

There Is No Free Lunch in RLVR: Every pass@1 Gain Has a pass@k Price

Reinforcement learning with verifiable rewards (RLVR) makes a language model solve more problems on the first

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

How a group that cannot communicate knows itself by its enemy

Teams that cannot communicate seem stuck with what each member sees. We show that their opponent can tell them

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

自然言語処理大規模言語モデルテキストマルチモーダル強化学習

CaPtain: A Pipeline for Post-Training Code-as-Policy Robot Agents

Across coding, web, and tool-use tasks, language-model agents are now routinely trained with the same recipe:

用途: 技術検証・論文読解補助
難易度: Easy
コスト: High

Probing Spatial State Representations in Visual Foundation Models

An observer sees only part of the surrounding world, yet must keep track of what lies beyond the current view.

用途: 技術検証・論文読解補助
難易度: Easy
コスト: Low

When Requirements Interact: Scaling Long-Horizon Coding via Interdependent Specifications

To succeed in real-world long-horizon software development, coding agents must satisfy multi-dimensional task

MLOpsパイプライン構築生成強化学習

用途: 生成
難易度: Easy
コスト: High

On the Origin and Evolution of Training-Inference Mismatch in LLM Reinforcement Learning

On-policy reinforcement learning for large language models assumes that training responses are sampled from th

用途: 技術検証・論文読解補助
難易度: Easy
コスト: High

強化学習方策勾配 (PPO / A3C)生成テキスト

MemSteer: Turning User Histories into Supervision for Agent Memory Optimization

An effective agent memory system should adapt to a user's history and task requirements. Human-annotated super

用途: 生成
難易度: Easy
コスト: Low

Training Foundation Models with a Single Algorithm

Foundation models are typically trained in stages: pretraining, midtraining, supervised fine-tuning (SFT), and

自然言語処理RAG生成強化学習

用途: 生成
難易度: Easy
コスト: High

Online Dense Reinforcement Learning with Planning-Grounding Agent for UI Navigation on Live IOS Apps

While frontier vision-language models demonstrate strong capabilities in user interface (UI) navigation, their

深層学習軽量化・量子化異常検知画像テキスト

用途: 異常検知
難易度: Easy
コスト: High

githubGitHubあり2026-10-29

Gymnasium — A standard API for single-agent reinforcement learning environments, with popular reference environments and related utilities (formerly Gym)

Gymnasiumは、シングルエージェントRLの疑似環境を提供するAPIです。

用途: 疑似環境を提供する
難易度: Easy
コスト: Medium

品質予測/異常検知強化学習方策勾配 (PPO / A3C)

CausalGrade: Evidence-Guided Causal Graph Scoring with Jev

Evaluating the structural quality of candidate causal graphs when the true causal structure is unknown is an i

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

コンピュータビジョンセグメンテーション生成強化学習

Estimating Implicit Expected Log Density Ratios

When exact probabilities cannot be computed, \emph{density ratio estimation (DRE)} is a key subroutine in the

用途: 生成
難易度: Hard
コスト: High

Principal-agent hypothesis testing under adaptive experimentation

Principal-agent hypothesis testing studies statistical contracts in which a principal rewards an agent accordi

説明可能強化学習

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Simultaneous Gradient Learning in First-Price Auctions

We study equilibrium learning in discretized Bayesian games, focusing on first-price auctions as a testbed for

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Game-Theoretic Foundations for Undiscounted Bisimulation Metrics

Bisimulation metrics offer a principled notion of task-relevant state similarity in reinforcement learning, bu

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

huggingfaceHugging Faceあり2026-10-28

Scaling Lean Theorem Proving for Real-World Code Verification

We present EasyProver, a Lean proving system built from recursive agents with isolated contexts, asynchronous

コンピュータビジョンセグメンテーションテキスト強化学習

用途: 技術検証・論文読解補助
難易度: Easy
コスト: High

Joint Distributional Copula Regression for Insurance Claim Frequency and Severity via Boosting

In this paper, we introduce a model-based boosting approach for fitting structured additive distributional cop

説明可能強化学習モデルベース回帰

用途: 回帰
難易度: Hard
コスト: Medium

Autonomous fixed-wing interception using reinforcement learning

Medium altitude, long endurance unmanned aerial systems such as Shahed-type strike drones are an unanswered th

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Multi-Agent Reinforcement Learning for Global Climate Policy Cooperation

International climate policy requires coordination among countries whose mitigation decisions impose local cos

自然言語処理RAG強化学習

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Low

Strategyproof Deep Reinforcement Learning for Online Resource Allocation

Online resource allocation faces two coupled challenges: (1) allocating scarce resources efficiently under unc

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

Characterizing Maximum Nash Welfare: Envy-Freeness and Proportionality

When allocating indivisible goods, the maximum Nash welfare rule is known to be the unique welfarist rule sati

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Impossibility of No-Regret in Multi-Agent Bidding

We consider $T$ repeated second-price auctions across $n$ pacing bidders with i.i.d. item arrivals, and show t

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

強化学習モデルフリー (DQN / SAC)テキスト

Tight Characterizations of AMM Mechanism Design under Weak Efficiency

Miner extractable value (MEV) poses a persistent challenge for blockchain ecosystems. It arises from the fact

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Sybil-Proofness of Generalized Adaptive Clinching and Implications for Blockchain Mechanism Design

The adaptive clinching auction of Dobzinski, Lavi, and Nisan (FOCS'08) is a cornerstone of mechanism design wi

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Learning-Augmented Coordination Mechanisms

The inefficiency of decentralized resource allocation is classically measured by the Price of Anarchy (PoA) an

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

huggingfaceGitHubありHugging Faceあり2026-10-27

Multi-Agent Relay Thinking

Existing multi-agent collaboration (MAC) methods have primarily explored a group thinking collaboration topolo

用途: 技術検証・論文読解補助
難易度: Easy
コスト: Medium

githubGitHubあり2026-10-27

Book-Mathematical-Foundation-of-Reinforcement-Learning — This is the homepage of a new book entitled "Mathematical Foundations of Reinforcement Learning."

Mathematical Foundations of Reinforcement Learningは、ディープラーニングにおける推論力学習の数学的基礎を網羅している。

用途: ディープラーニングに関する本書の制作
難易度: Easy
コスト: Medium

A Dynamic Decomposition Strategy for the MOEA/D With Proven Performance Guarantees

The MOEA/D is one of the most successful multi-objective evolutionary algorithms. It operates by decomposing t

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Bayesian Replicator Dynamics and the Inversion of Strict Dominance in Bimatrix Games

Replicator dynamics, the central model of evolutionary game theory, describe populations evolving under select

強化学習方策勾配 (PPO / A3C)分類

用途: 分類
難易度: Hard
コスト: Low

Formation of international environmental agreements and payoff allocation

Dealing with climate change requires that all countries engage in costly efforts to reduce their emissions. Re

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Trim: Fee Level Matters More Than Skew in Lightning Routing

Lightning routing nodes lend their channel balance for a posted fee. Operators' tools move that fee with the b

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

品質予測/異常検知強化学習方策勾配 (PPO / A3C)

BROCADE: Broker-based Contract Design for the Multi-Agency Model

We study welfare maximization in a contract model with $m$ principals who privately value a public project's s

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

huggingfaceHugging Faceあり2026-10-26

CEPO: A General Framework for Improving Prompt Optimization in Multi-Agent Systems

Multi-agent systems (MAS) require both effective coordination and specialized task execution. Existing prompt

用途: 技術検証・論文読解補助
難易度: Easy
コスト: Medium

githubGitHubあり2026-10-26

stable-baselines3 — PyTorch version of Stable Baselines, reliable implementations of reinforcement learning algorithms.

このリポジトリでは、LLMベースのエージェントアプリケーションのための強化学習の橋渡しを提供しています。

用途: 強化学習を簡素化させる橋渡し
難易度: Easy
コスト: High

githubGitHubあり2026-10-25

FinGPT — FinGPT: Open-Source Financial Large Language Models! Revolutionize 🔥 We release the trained model on HuggingFace.

このリポジトリでは、Lecture Learning Modelsに対してReinforcement Learningを実行するライブラリを提供しています。

自然言語処理大規模言語モデルテキスト

用途: 可搬性のあるReinforcement Learning
難易度: Easy
コスト: High

arxivPaper only2026-10-24

Sensorimotor behaviour from an untrained brain-and-nerve-cord connectome

Connectomes now map the neurons and synapses of an adult fly's brain and nerve cord, but how much of behaviour

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

arxivPaper only2026-10-24

The Conditional Value of Costly Information in Collective-Risk Dilemmas

Many high-stakes collective problems, e.g., climate mitigation and pandemic response, share a structure in whi

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

arxivPaper only2026-10-24

More Meters or More Prices? One- vs. Multi-dimensional Fee Mechanisms on Ethereum

Permissionless blockchains use transaction fee mechanisms to allocate the resources that transactions consume,

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

arxivPaper only2026-10-23

Capacity Modification for Matching with Couples: An Integer Programming and Computational Study

We study capacity modification in the Hospitals/Residents problem with Couples, in which hospital capacities a

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

githubGitHubあり2026-10-23

easy-rl — 强化学习中文教程（蘑菇书🍄），在线阅读地址：https://datawhalechina.github.io/easy-rl/

强化学习中文教程（蘑菇书🍄），在线阅读地址：https://datawhalechina.github.io/easy-rl/

用途: 実装・検証基盤
難易度: Easy
コスト: Medium

Differentiable Combinatorial Auction Design: Optimizing Revenue-Welfare-Fairness Frontiers With Graph Attention Networks

Designing optimal auctions for allocating public infrastructure-such as airport slots and energy resources-is

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Randomization to Balance Performance under Nominal and Compromised Communication

It is well-known that under nominal communication, where agents are able to share information with one another

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

A Resource Allocation Tournament

We introduce and analyze a round-robin simultaneous resource allocation contest, motivated by sports leagues a

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Identifying level-$k$ reasoning in repeated games: Strategies, beliefs, and cognitive ability

In this paper, we identify level-$k$ reasoning in repeated games that operates at the level of a supergame str

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Matching Tasks to Agents: Mechanism Design for Multi-Agent Contracts

We study task assignment under adverse selection and moral hazard. In the single-task setting, agents privatel

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Fast Iterative Combinatorial Auctions via Deep Reinforcement Learning

Iterative combinatorial auctions are used in spectrum allocation, transportation, and industrial procurement,

深層学習RNN / LSTM強化学習

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Computing the Nucleon of Matching Games

The nucleon is the multiplicative counterpart of the nucleolus: it lexicographically maximizes the fraction of

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Formalizing Cooperation and Competition: An AI-Assisted Mathematical Addendum to a 2018 Planning Thesis

We present an AI-assisted mathematical addendum to Strategic Trip Planning: Striking a Balance Between Competi

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

GSP Autobidding Equilibria with Few Advertisers or Slots

In sponsored-shopping auctions, one advertiser may win several positions, and a product that is never displaye

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Correlated Quantal Response Equilibrium: A New Equilibrium Notion for Human-AI Strategic Interactions

The modeling of the behavior of agents in strategic environments and the prediction of the resulting outcomes

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

huggingfaceHugging Faceあり2026-10-21

LongZero: Enhancing Long-Context Reasoning in LLMs from Zero Data

Reinforcement learning (RL) is an effective way to strengthen the long-context reasoning of large language mod

自然言語処理大規模言語モデル生成テキスト強化学習

用途: 生成
難易度: Easy
コスト: High

Is Anyone Enforcing the Cartel? What Reinforcement Learning Teaches LLM Agents in Markets

Evaluations of large language model (LLM) agents report that pricing agents collude at others' expense. Yet co

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

Subjective Valuations Make It Hard to be Fair: The Complexity of Resource Allocation to Groups of Agents

We study the parameterized complexity of finding an EF1 allocation of indivisible items in the group-sharing s

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Best-of-Both Worlds Proportionality in Structured Committee Elections

We study randomized committee elections with additive preferences, where we seek best-of-both-worlds proportio

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Fixed Sets and the Hardness of Computing Correlated Equilibria

We show that computing an approximate correlated equilibrium of a polytope game is PPAD-complete, and thus pol

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Extending Mechanisms

Given an incentive-compatible mechanism for a single agent on a restricted domain of valuations, we consider i

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

githubGitHubあり2026-10-20

Practical_RL — A course in reinforcement learning in the wild

A course in reinforcement learning in the wild

用途: 実装・検証基盤
難易度: Easy
コスト: Medium

Sector-Ring Linear Reservoirs: Sector-Wise Decay-Timescale Allocation With State Reuse Across Candidate Settings

In reservoir computing, performance depends on whether the fixed dynamics of the reservoir suit the task, and

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

Fundamental Impossibility Results in AI Safety: Trading Off between Desiderata

Is there a fundamental tradeoff between desiderata for safe AI systems, such as harmlessness and helpfulness?

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Sustainable and Envy-free Multi-Period Allocation

Allocating renewable resources fairly requires accounting for how present decisions affect future availability

品質予測/異常検知強化学習

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Hardness Amplification for Approximate Nash Equilibrium with Constantly Many Players

For every fixed $\varepsilon\in(0,1)$ and $η\in[0,1)$, we prove that computing an $(\varepsilon,η)$-weak appro

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

arxivPaper only2026-10-18

Breaking the 1/3 Barrier for Approximate Nash Equilibria in Symmetric Bimatrix Games

We give the first unconditional family of polynomial-time algorithms that breaks the $1/3$ approximation barri

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

arxivPaper only2026-10-18

Improved Approximation and Tight Competition for Simple Auctions

We study simple auctions for one additive buyer whose values for $m$ items are independent. Write $\mathrm{REV

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

arxivPaper only2026-10-17

Cooperation Without Shared Rewards via Reciprocal Q-Augmentation

General-sum games are more challenging than fully cooperative or adversarial settings because agents may decei

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

arxivPaper only2026-10-17

Blockchain Participation Games with Upfront Contribution Commitment

Blockchain participation games combine costly entry, random participation selection, and threshold success int

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Uniform-Price Mechanism Design for Automated Market Makers

Automated market makers (AMMs) are central to decentralized exchanges, but their reserve curves make mechanism

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Contracting with an Agent under Model Mismatch

In the classical principal-agent setting, the principal delegates some task to the agent, who receives a payme

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Bayesian Learning of Parametric Mean Field Games with Common Noise

We study the inverse problem of recovering unknown parameters governing the dynamics and rewards of Mean Field

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Trembling toward Duverger: Convergence and Cycles in Synchronous Plurality Dynamics

Duverger's law states that plurality elections tend to concentrate support on two candidates. Pivotal-voter mo

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Proportionality in Ranking Compression

Motivated by applications in reinforcement learning from human feedback, we consider a setting where a given c

強化学習RLHF

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

githubGitHubあり2026-10-16

rllm — Democratizing Reinforcement Learning for LLMs

このリポジトリでは、AIエンジニアリングのためのリソースを提供しています。

用途: AIエンジニアリング
難易度: Easy
コスト: High

arxivPaper only2026-10-15

Auditable and Deployable Multi-Agent Motion Planning via Posetal Games from Legal Guidelines

When robots are deployed in the real-world, they must interact with other agents and accomplish their own goal

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

arxivPaper only2026-10-15

Learning Rewards in Strategic Games

Understanding the objectives that drive agents' behavior is crucial in strategic environments, where multiple

自然言語処理ファインチューニング強化学習

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

arxivPaper only2026-10-15

Approximating the Value of Time-Bounded Weighted Timed Games

The Value Problem for weighted timed games, an extension of timed automata with weights and adversity, consist

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

githubGitHubあり2026-10-15

vowpal_wabbit — Vowpal Wabbit is a machine learning system which pushes the frontier of machine learning with techniques such as online, hashing, allreduce, reductions, learning2search, active, and interactive learning.

Vowpal Wabbitは、機械学習を進歩させるためのオンライン学習、ハッシュ、reduceなどの強力なアルゴリズムを含むシステムです。その結果、さまざまな問題に応じて、高品質な解決策を提供できます。

強化学習テキスト

用途: 強い機械学習アルゴリズムを実行し複雑な問題を解決するためのシステム
難易度: Easy
コスト: Medium

An autocomplete model for elections

We consider two linked problems for ranked choice elections: building a compact summary of a preference profil

強化学習生成

用途: 生成
難易度: Hard
コスト: Medium

Auction-based allocation of routes in capacity-constrained networks

We study the route-set allocation problem between end nodes in capacity-constrained flow networks with transfe

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Accord: Schelling-Point Arbitration as a Composable Solana Primitive

Accord is an arbitration oracle on Solana. Any Solana program can create an application specific court (i.e.

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

The Two-Level Poset Game is PSPACE-Complete

Poset games form a fundamental class of combinatorial games that includes classical games such as Nim and Chom

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

githubGitHubあり2026-10-14

PufferLib — Puffing up reinforcement learning

用途: 強化学習用ライブラリ
難易度: Easy
コスト: Medium

Maximin share allocations for two types of items

We study the fair allocation of indivisible items among $n$ agents with additive valuations (or costs) using t

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

The Geography of (Concurrent) Block Proposing

The resilience and censorship resistance of a blockchain depend on where the validators and proposers of that

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Falsification-proof non-market allocation mechanisms

Relying on manipulable characteristics to target the allocation of resources is fraught with difficulties. Nai

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

説明可能強化学習モデルフリー (DQN / SAC)

Strategic Multi-Agent Resource Allocation via Statistical Contracts

We study a class of strategic hypothesis testing problems in which a principal aims to allocate resources amon

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

When Does Opponent Modeling Pay Off? Spatial Inference Stabilises Reinforcement Learning under Partial Observability

Deep reinforcement learning agents in partially observable competitive games must act on beliefs about an unse

深層学習Transformer強化学習

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

Truthful Two-Facility Location with Candidate Locations in Metric Spaces

We study a truthful facility location problem in which agents have private positions in a metric space and kno

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Certifying Capacity and Gaps of the Rank-Dependent Gain-Sharing Framework for Ranking

The randomized primal-dual framework of Devanur, Jain, and Kleinberg (SODA 2013) splits each unit matching gai

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Time Window Optimization for Motor Imagery Classification via Temporal Structure Embedded Reinforcement Learning

Online motor imagery-based brain-computer interfaces (MI-BCIs) typically decode continuous EEG using fixed-len

深層学習Transformer分類画像強化学習

用途: 分類
難易度: Hard
コスト: High

Unit-Demand Pricing for Min-Buying Customers

We study the unit-demand pricing problem under \textit{min-buying} behavior, where customers purchase the lowe

強化学習テキスト

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

Dynamic Run-Out Pricing for Perishable Goods

I study a dynamic pricing problem in which a seller with limited inventory must sequentially price a perishabl

説明可能強化学習方策勾配 (PPO / A3C)

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

An Introduction to Contract Theory for Computer Scientists

This survey introduces contract design to computer scientists: using outcome-based payments to incentivize eff

強化学習方策勾配 (PPO / A3C)分類

用途: 分類
難易度: Hard
コスト: Low

Digital-Twin-Referenced Belief-Guided Q-Learning for Jammer Identification and Beam-Power Control in a mmWave Model

Next-generation wireless networks are expected to support high data rates and reliable connectivity while meet

深層学習Transformer検出生成強化学習

用途: 検出
難易度: Hard
コスト: Medium

githubGitHubあり2026-10-09

Awesome-Process-Reward-Models — A comprehensive collection of process reward models.

医療では、イメージャは単に画像を解釈するのではなく、複数の画像を比較して診断を行うことが多い。しかし、現在の技術ではこのような比較を行うことは困難であるため、メドリコのDBというデータセットを利用することで、医療の比較推

強化学習RLHF

用途: 医療における画像の比較
難易度: Easy
コスト: Medium

githubGitHubあり2026-10-09

bullet3 — Bullet Physics SDK: real-time collision detection and multi-physics simulation for VR, games, visual effects, robotics, machine learning etc.

Bullet Physics SDK: real-time collision detection and multi-physics simulation for VR, games, visual effects,

強化学習検出画像

用途: 検出
難易度: Easy
コスト: Medium

arxivPaper only2026-10-08

Learning to Orchestrate Evolutionary Search: Progression-Aware Deep Reinforcement Learning for Dynamic DE-CMA-ES Coordination in Optimization and Structural Model Updating

Solving high-dimensional structural model updating problems requires an algorithm capable of navigating comple

自然言語処理ファインチューニング生成強化学習

用途: 生成
難易度: Hard
コスト: High

arxivPaper only2026-10-06

Common-Mode Errors Limit Low-Timestep Deep Spiking Q-Networks

Spiking neural networks (SNNs) offer sparse and event-driven computation, making them attractive for energy-co

コンピュータビジョン物体検出強化学習

用途: 技術検証・論文読解補助
難易度: Hard
コスト: Medium

arxivPaper only2026-10-05

Visual Swarm Navigation via Deep Reinforcement Learning and Evolutionary Hybrid Design

Swarm robotics presents a robust and cost-effective paradigm for advanced automation in complex, dynamic envir

深層学習軽量化・量子化生成画像強化学習

用途: 生成
難易度: Hard
コスト: Low

arxivPaper only2026-10-01

Continual Reinforcement Learning with Neuroevolution

Despite many studies about causes and remedies of plasticity loss in Reinforcement Learning (RL) under continu

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

arxivPaper only2026-10-01

Inherited Learning in an Artificial Ecology: How Controls and Update Allocation Shape Benefits

Learning can improve an individual's behavior, yet a population risks losing that experience whenever individu

品質予測/異常検知強化学習方策勾配 (PPO / A3C)生成

用途: 生成
難易度: Hard
コスト: Medium

arxivPaper only2026-09-30

From Search to Signal: Online Post-Training in Automatic Heuristic Design

Large language model (LLM)-based automatic heuristic design (AHD) iteratively proposes and refines heuristics,

品質予測/異常検知自然言語処理大規模言語モデル生成テキスト強化学習

用途: 生成
難易度: Hard
コスト: High

arxivPaper only2026-09-30

T-Router: Learning Thalamic Routing for Reasoning with Parameter-Efficient Reinforcement Learning

Parameter-efficient reinforcement learning aims to improve reasoning with a compact trainable interface to a p

深層学習RNN / LSTMテキスト強化学習

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

arxivPaper only2026-09-29

Brain-SAD: A Brain-Inspired Safe Autonomous Driving Control Framework with Dynamic Fear-Oriented Constraint on Dual-Policy

Constrained Reinforcement Learning has recently gained increasing attention in the field of Safe Autonomous Dr

数学・理論最適化強化学習

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

arxivPaper only2026-09-29

Bias Robustness of Genetic Multi-Criteria Sorting versus Learned Rankers under Proxy Skew and Label Contamination

Learning-to-rank (LTR) models can align rankings with undesignated signals when those signals appear as featur

表形式向きCPUで試しやすい強化学習方策勾配 (PPO / A3C)

用途: 技術検証・論文読解補助
難易度: Hard
コスト: High

arxivPaper only2026-09-28

Massively Parallel Reinforcement Learning with a Chaotic Reconfigurable Clockless Chip

Hardware accelerators based on physical dynamical systems offer an attractive route toward energy-efficient re

深層学習軽量化・量子化生成強化学習

用途: 生成
難易度: Hard
コスト: Medium