Martin Riedmiller

収録論文 44本 ・ フィジカルAI/ロボット学習

※arXiv著者名で収集。同姓同名の別人の論文が含まれる場合があります。

論文

• EXIMO: VLMによるVLAポリシー探索のガイド
  VLA/強化学習2026/8/20

  VLAポリシーの効率的なファインチューニング手法EXIMOを提案。VLMをプランナーとして使い、長期的なタスクを分解してデータ収集し、模倣学習とオフポリシーRLで最適化する。

• ゲーム開始：言語モデルを強化学習の実験者として活用する
  VLA2024/9/1

  VLMを用いて強化学習の実験サイクルを自動化し、タスク提案・スキル分解・カリキュラム構築を行うエージェントアーキテクチャを提案した。

• DemoStart: 実演主導の自動カリキュラムによる多指ロボットのsim-to-real
  sim2real2024/9/1

  シミュレーションでの少数の実演と疎な報酬のみから、3本指ロボットハンドの複雑なマニピュレーション行動を自動カリキュラム強化学習で獲得し、ゼロショットで実機に転移させる手法を提案した。

• 自己中心視覚と深層強化学習によるロボットサッカーの学習
  マルチエージェント強化学習2024/5/1

  自己中心的なRGB視覚のみを入力とし、関節レベルの行動を出力するマルチエージェントロボットサッカーの方策を深層強化学習で訓練し、シミュレーションから実機へ転移することに成功した。

• Offline Actor-Critic Reinforcement Learning Scales to Large Models
  2024/2/1
• Real-World Fluid Directed Rigid Body Control via Deep Reinforcement Learning
  2024/2/1
• Less is more -- the Dispatcher/ Executor principle for multi-task Reinforcement Learning
  2023/12/1
• Mastering Stacking of Diverse Shapes with Large-Scale Iterative Reinforcement Learning on Real Robots
  2023/12/1
• Replay across Experiments: A Natural Extension of Off-Policy RL
  2023/11/1
• Equivariant Data Augmentation for Generalization in Offline Reinforcement Learning
  2023/9/1
• Policy composition in reinforcement learning via multi-objective policy optimization
  2023/8/29
• Real Robot Challenge 2022: Learning Dexterous Manipulation from Offline Data in the Real World
  2023/8/1
• Towards A Unified Agent with Foundation Models
  2023/7/1
• RoboCat: A Self-Improving Generalist Agent for Robotic Manipulation
  2023/6/1
• A Generalist Dynamics Model for Control
  2023/5/1
• Leveraging Jumpy Models for Planning and Fast Learning in Robotic Domains
  2023/2/1
• SkillS: Adaptive Skill Sequencing for Efficient Temporally-Extended Exploration
  2022/11/1
• Solving Continuous Control via Q-learning
  2022/10/1
• MO2: Model-Based Offline Options
  2022/9/1
• Revisiting Gaussian mixture critics in off-policy reinforcement learning: a sample-based approach
  2022/4/21
• Is Bang-Bang Control All You Need? Solving Continuous Control with Bernoulli Policies
  2021/11/1
• Beyond Pick-and-Place: Tackling Robotic Stacking of Diverse Shapes
  2021/10/1
• Evaluating model-based planning and planner amortization for continuous control
  2021/10/1
• Is Curiosity All You Need? On the Utility of Emergent Behaviours from Curious Exploration
  2021/9/1
• On Multi-objective Policy Optimization as a Tool for Reinforcement Learning: Case Studies in Offline RL and Finetuning
  2021/6/1
• Representation Matters: Improving Perception and Exploration for Robotics
  2020/11/1
• "What, not how": Solving an under-actuated insertion task from scratch
  2020/10/1
• Towards General and Autonomous Learning of Core Skills: A Case Study in Locomotion
  2020/8/1
• Data-efficient Hindsight Off-policy Option Learning
  2020/7/1
• Simple Sensor Intentions for Exploration
  2020/5/1
• A Distributional View on Multi-Objective Policy Optimization
  2020/5/1
• Keep Doing What Worked: Behavioral Modelling Priors for Offline Reinforcement Learning
  2020/2/1
• Continuous-Discrete Reinforcement Learning for Hybrid Control in Robotics
  2020/1/1
• Imagined Value Gradients: Model-Based Policy Optimization with Transferable Latent Dynamics Models
  2019/10/1
• Compositional Transfer in Hierarchical Reinforcement Learning
  2019/6/1
• Simultaneously Learning Vision and Feature-based Control Policies for Real-world Ball-in-a-Cup
  2019/2/1
• Self-supervised Learning of Image Embedding for Continuous Control
  2019/1/1
• Maximum a Posteriori Policy Optimisation
  2018/6/1
• Learning by Playing - Solving Sparse Reward Tasks from Scratch
  2018/2/1
• PVEs: Position-Velocity Encoders for Unsupervised Learning of Structured State Representations
  2017/5/1
• Data-efficient Deep Reinforcement Learning for Dexterous Manipulation
  2017/4/1
• Learning and Transfer of Modulated Locomotor Controllers
  2016/10/1
• Multimodal Deep Learning for Robust RGB-D Object Recognition
  2015/7/1
• Playing Atari with Deep Reinforcement Learning
  2015/2/24