RT-Trajectory: Robotic Task Generalization via Hindsight Trajectory Sketches
Comments Evaluation videos can be found at https://rt-trajectory.github.io/
作者
Chelsea Finn
Robotics / Machine Learning
Open-World Object Manipulation using Pre-trained Vision-Language Models
Comments Accepted at the 7th Conference on Robot Learning (CoRL 2023)
Just Ask for Calibration: Strategies for Eliciting Calibrated Confidence Scores from Language Models Fine-Tuned with Human Feedback
Comments EMNLP 2023 Camera Ready
Robot Fine-Tuning Made Easy: Pre-Training Rewards and Policies for Autonomous Real-World Reinforcement Learning
Disentanglement via Latent Quantization
Comments NeurIPS 2023 camera-ready. 26 pages, 15 figures. Code available at https://github.com/kylehkhsu/latent_quantization
Meta-Learning Online Adaptation of Language Models
Comments EMNLP 2023 Camera Ready
An Emulator for Fine-Tuning Large Language Models using Small Language Models
Q-Transformer: Scalable Offline Reinforcement Learning via Autoregressive Q-Functions
Comments See website at https://qtransformer.github.io
Zero-Shot Robotic Manipulation with Pretrained Image-Editing Diffusion Models
Comments 22 pages, 8 figures
Offline Retraining for Online RL: Decoupled Policy Learning to Mitigate Exploration Bias
RoboCLIP: One Demonstration is Enough to Learn Robot Policies
Bitrate-Constrained DRO: Beyond Worst Case Robustness To Unknown Group Shifts
Journal ref ICLR 2023
Multi-Domain Long-Tailed Learning by Augmenting Disentangled Representations
Comments Accepted by TMLR
Permutation Equivariant Neural Functionals
Comments To appear in Neural Information Processing Systems (NeurIPS), 2023
Pre-Training for Robots: Offline RL Enables Learning New Tasks from a Handful of Trials
Robot Parkour Learning
Comments CoRL 2023 (Oral). Project website at https://robot-parkour.github.io
RT-1: Robotics Transformer for Real-World Control at Scale
Comments See website at robotics-transformer1.github.io
Self-Destructing Models: Increasing the Costs of Harmful Dual Uses of Foundation Models
Comments v1 Presented at the First Workshop of Pre-training: Perspectives, Pitfalls, and Paths Forward (ICML, 2022) and New Frontiers in Adversarial Machine Learning Workshop (ICML, 2022); v2 Presented at the Sixth AAAI/ACM Conference on AI, Ethics, and Society (AIES, 2023)
RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control
Comments Website: https://robotics-transformer.github.io/
Waypoint-Based Imitation Learning for Robotic Manipulation
Comments The first two authors contributed equally
Contrastive Example-Based Control
Comments This is an updated version of a manuscript that originally appeared at L4DC 2023. The project website is here https://sites.google.com/view/laeo-rl
Journal ref Proceedings of The 5th Annual Learning for Dynamics and Control Conference, PMLR 211:155-169, 2023
DetectGPT: Zero-Shot Machine-Generated Text Detection using Probability Curvature
Comments ICML 2023
Giving Robots a Hand: Learning Generalizable Manipulation with Eye-in-Hand Human Video Demonstrations
Comments 21 pages, 7 figures, project webpage at https://giving-robots-a-hand.github.io/
Polybot: Training One Policy Across Robots While Embracing Variability
Comments 17 pages, 11 figures
Decomposing the Generalization Gap in Imitation Learning for Visual Robotic Manipulation
Comments Project webpage at https://sites.google.com/view/generalization-gap
Train Offline, Test Online: A Real Robot Learning Benchmark
Comments Accepted to ICRA 2023
Supervised Pretraining Can Learn In-Context Reinforcement Learning
Confidence-Based Model Selection: When to Take Shortcuts for Subpopulation Shifts
Comments 15 pages, 5 figures
Simple Embodied Language Learning as a Byproduct of Meta-Reinforcement Learning
Comments International Conference on Machine Learning (ICML), 2023
Surgical Fine-Tuning Improves Adaptation to Distribution Shifts
Comments ICLR 2023
