Vikash Kumar

収録論文 54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

• VBVR-Pro: ネイティブ視覚推論のためのスケーラブルで検証可能なスイート
  視覚推論2026/8/26

  視覚生成を推論の媒体として扱うネイティブ視覚推論の研究を進めるため、300の手続き生成タスクからなる閉ループテストベッドVBVR-Proを導入し、検証可能な報酬スコアラと制御されたモダリティ比較を提供する。

• MyoChallenge 2025: 人間の運動知能のための新しいベンチマーク
  運動制御2026/5/1

  NeurIPS 2025で開催されたMyoChallengeは、高忠実度の筋骨格モデルと物理シミュレーションを用いて、卓球ラリーとサッカーペナルティキックの2つのタスクで運動制御の知能を競うベンチマークを確立した。

• 超大規模動画推論スイート
  動画推論2026/2/1

  200種類の推論タスクと100万本以上の動画からなる大規模データセットVBVRと、ルールベースで検証可能な評価ベンチマークVBVR-Benchを構築し、動画推論のスケーリング則と未見タスクへの汎化の兆候を示した。

• ニューラル運動計画のためのカスケード拡散モデル
  運動計画2025/5/1

  拡散ポリシーを用いて複雑な環境での衝突のない大域的な運動計画を学習するカスケード階層モデルを提案し、ナビゲーションやマニピュレーションで既存手法を約5%上回る性能を示した。

• 意味的に制御可能な拡張による汎化可能なロボット学習
  sim2real2024/9/1

  画像テキスト生成モデルをデータ拡張に活用し、ロボット操作データを多様化することで、未見の実環境への汎化を可能にする学習フレームワークを提案した。

• Towards Generalizable Zero-Shot Manipulation via Translating Human Interaction Plans
  2023/12/1
• Open X-Embodiment: Robotic Learning Datasets and RT-X Models
  2023/10/1
• RoboHive: A Unified Framework for Robot Learning
  2023/10/1
• REBOOT: Reuse Data for Bootstrapping Efficient Real-World Dexterous Manipulation
  2023/9/1
• Natural and Robust Walking using Reinforcement Learning without Demonstrations in High-Dimensional Musculoskeletal Models
  2023/9/1
• MoDem-V2: Visuo-Motor World Models for Real-World Robot Manipulation
  2023/9/1
• MyoDex: A Generalizable Prior for Dexterous Manipulation
  2023/9/1
• RoboAgent: Generalization and Efficiency in Robot Manipulation via Semantic Augmentations and Action Chunking
  2023/9/1
• SAR: Generalization of Physiological Agility and Dexterity via Synergistic Action Representation
  2023/7/1
• LIV: Language-Image Representations and Rewards for Robotic Control
  2023/6/1
• Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware
  2023/4/1
• GenAug: Retargeting behaviors to unseen situations via Generative Augmentation
  2023/2/1
• Zero-Shot Robot Manipulation from Passive Human Videos
  2023/2/1
• Cross-Domain Transfer via Semantic Skill Imitation
  2022/12/1
• CACTI: A Framework for Scalable Multi-Task Multi-Scene Visual Imitation Learning
  2022/12/1
• MoDem: Accelerating Visual Model-Based Reinforcement Learning with Demonstrations
  2022/12/1
• Dexterous Manipulation from Images: Autonomous Real-World RL via Substep Guidance
  2022/12/1
• Visual Dexterity: In-Hand Reorientation of Novel and Complex Object Shapes
  2022/11/1
• All the Feels: A dexterous hand with large-area tactile sensing
  2022/10/1
• Real World Offline Reinforcement Learning with Realistic Data Source
  2022/10/1
• VIP: Towards Universal Visual Reward and Representation via Value-Implicit Pre-Training
  2022/10/1
• Learning Dexterous Manipulation from Exemplar Object Trajectories and Pre-Grasps
  2022/9/1
• MyoSuite -- A contact-rich simulation suite for musculoskeletal motor control
  2022/5/1
• Can Foundation Models Perform Zero-Shot Task Specification For Robot Manipulation?
  2022/4/1
• Curiosity Driven Self-supervised Tactile Exploration of Unknown Objects
  2022/4/1
• RB2: Robotic Manipulation Benchmarking with a Twist
  2022/3/1
• Policy Architectures for Compositional Generalization in Control
  2022/3/1
• R3M: A Universal Visual Representation for Robot Manipulation
  2022/3/1
• RRL: Resnet as representation for Reinforcement Learning
  2021/7/1
• Reset-Free Reinforcement Learning via Multi-Task Learning: Learning Dexterous Manipulation Behaviors without Human Intervention
  2021/4/1
• The Ingredients of Real-World Robotic Reinforcement Learning
  2020/4/1
• A Game Theoretic Framework for Model Based Reinforcement Learning
  2020/4/1
• Emergent Real-World Robotic Skills via Unsupervised Off-Policy Reinforcement Learning
  2020/4/1
• Benchmarking In-Hand Manipulation
  2020/1/1
• Relay Policy Learning: Solving Long-Horizon Tasks via Imitation and Reinforcement Learning
  2019/10/1
• Deep Dynamics Models for Learning Dexterous Manipulation
  2019/9/1
• ROBEL: Robotics Benchmarks for Learning with Low-Cost Robots
  2019/9/1
• Multi-Agent Manipulation via Locomotion using Hierarchical Sim2Real
  2019/8/1
• Dynamics-Aware Unsupervised Discovery of Skills
  2019/7/1
• Learning Latent Plans from Play
  2019/3/1
• Soft Actor-Critic Algorithms and Applications
  2018/12/1
• Dexterous Manipulation with Deep Reinforcement Learning: Efficient, General, and Low-Cost
  2018/10/1
• Time Reversal as Self-Supervision
  2018/10/1
• Multi-Goal Reinforcement Learning: Challenging Robotics Environments and Request for Research
  2018/2/1
• Divide-and-Conquer Reinforcement Learning
  2017/11/1
• Domain Randomization and Generative Models for Robotic Grasping
  2017/10/1
• Learning Complex Dexterous Manipulation with Deep Reinforcement Learning and Demonstrations
  2017/9/1
• Pneumatic Modelling for Adroit Manipulation Platform
  2017/3/1
• Learning Dexterous Manipulation Policies from Experience and Imitation
  2016/11/1