Policy Generalization

Policy generalization in artificial intelligence focuses on developing algorithms and models that can adapt and perform well on unseen tasks or environments, going beyond the specific data used for training. Current research emphasizes techniques like active learning to efficiently fine-tune pre-trained policies, leveraging graph neural networks and transformers to capture complex relationships in robotics, and employing methods such as disentangled representations and invariant causal imitation learning to improve robustness and generalization across diverse scenarios. This research is crucial for creating more adaptable and reliable AI systems, with significant implications for fields like robotics, healthcare, and automated decision-making.

Papers

October 28, 2024

Identifying Selections for Unsupervised Subtask Discovery
Yiwen Qiu, Yujia Zheng, Kun Zhang
Reinforcement Learning View Selection Long Horizon Task Policy Generalization Subtask Assignment

October 13, 2024

Generalization of Compositional Tasks with Logical Specification via Implicit Planning
Duo Xu, Faramarz Fekri
Reinforcement Learning Strong Generalization Goal Conditioned Policy Compositional Task Policy Generalization

October 7, 2024

Active Fine-Tuning of Generalist Policies
Marco Bagatella, Jonas Hübotter, Georg Martius, Andreas Krause
Imitation Learning Fast Adaptation Policy Generalization Multi Task Policy Active Finetuning Multi Task Fine Tuning Policy Fine Tuning

September 26, 2024

Open Digital Rights Enforcement Framework (ODRE): from descriptive to enforceable policies
Andrea Cimmino, Juan Cano-Benito, Raúl García-Castro
Top Level Ontology Policy Generalization

August 2, 2024

HeteroMorpheus: Universal Control Based on Morphological Heterogeneity Modeling
YiFan Hao, Yang Yang, Junru Song, Wei Peng, Weien Zhou, Tingsong Jiang, Wen Yao
Robot Dynamic Morphological Differentiation Based Method Robot Morphology Policy Generalization Global Control

July 11, 2024

HACMan++: Spatially-Grounded Motion Primitives for Manipulation
Bowen Jiang, Yilin Wu, Wenxuan Zhou, Chris Paxton, David Held
Action Space Strategic Manipulation Motion Primitive Long Horizon Manipulation Task Policy Generalization

July 9, 2024

Using Large Language Models for Generating Smart Contracts for Health Insurance from Textual Policies
Inwon Kang, William Van Woensel, Oshani Seneviratne
Complex Process Smart Contract Health Policy Recommendation Policy Generalization Health Insurance

April 25, 2024

Offline Reinforcement Learning with Behavioral Supervisor Tuning
Padmanaba Srinivasan, William Knottenbelt
Offline Reinforcement Learning Policy Generalization Offline Dataset Human Supervisor

February 27, 2024

Agent-Pro: Learning to Evolve via Policy-Level Reflection and Optimization
Wenqi Zhang, Ke Tang, Hai Wu, Mengna Wang, Yongliang Shen, Guiyang Hou, Zeqi Tan, Peng Li, Yueting Zhuang, Weiming Lu
LeArning Abstract Optimization Purpose Agent Smith Policy OpTimization LLM Based Agent Policy Generalization Evolving Policy

February 5, 2024

Understanding What Affects the Generalization Gap in Visual Reinforcement Learning: Theory and Empirical Evidence
Jiafei Lyu, Le Wan, Xiu Li, Zongqing Lu
Theoretical Understanding Generalization Gap Visual Reinforcement Learning Empirical Evidence Policy Generalization

November 2, 2023

Invariant Causal Imitation Learning for Generalizable Policies
Ioana Bica, Daniel Jarrett, Mihaela van der Schaar
Causal Feature Policy Generalization Imitation Policy Qualitative Feature Causal Imitation

October 26, 2023

Large Language Models as Generalizable Policies for Embodied Tasks
Andrew Szot, Max Schwarzer, Harsh Agrawal, Bogdan Mazoure, Walter Talbott, Katherine Metcalf, Natalie Mackraz, Devon Hjelm, Alexander Toshev
Large Language Model Reinforcement Learning Policy Generalization Rearrangement Task Embodied Task

June 19, 2023

Effect-Invariant Mechanisms for Policy Generalization
Sorawit Saengkyongam, Niklas Pfister, Predrag Klasnja, Susan Murphy, Jonas Peters
Zero Shot Policy Generalization Invariant Causal

February 13, 2023

GenAug: Retargeting behaviors to unseen situations via Generative Augmentation
Zoey Chen, Sho Kiami, Abhishek Gupta, Vikash Kumar
Generative Model Strong Generalization Robot Learning Pre Trained Generative Model Behavior Style Policy Generalization

November 30, 2022

General policy mapping: online continual reinforcement learning inspired on the insect brain
Angel Yanguas-Gil, Sandeep Madireddy
Reinforcement Learning Algorithm Efficient Reinforcement Learning Continual Reinforcement Learning Lifelong Reinforcement Learning Policy Generalization Insect Machine Hybrid System Backward Transfer

November 8, 2022

Active Example Selection for In-Context Learning
Yiming Zhang, Shi Feng, Chenhao Tan
Large Language Model Language Model Context Learning Large Scale Language Model Context Example Policy Generalization

October 1, 2022

Zero-Shot Policy Transfer with Disentangled Task Representation of Meta-Reinforcement Learning
Zheng Wu, Yichen Xie, Wenzhao Lian, Changhao Wang, Yanjiang Guo, Jianyu Chen, Stefan Schaal, Masayoshi Tomizuka
Reinforcement Learning Zero Shot Meta Reinforcement Learning Compositional Task Policy Generalization

September 16, 2022

Towards A Unified Policy Abstraction Theory and Representation Learning Approach in Markov Decision Processes
Min Zhang, Hongyao Tang, Jianye Hao, Yan Zheng
Representation Learning Markov Decision Process Policy Generalization Policy Representation Expressive Policy Policy Architecture

June 27, 2022

Prompting Decision Transformer for Few-Shot Policy Generalization
Mengdi Xu, Yikang Shen, Shun Zhang, Yuchen Lu, Ding Zhao, Joshua B. Tenenbaum, Chuang Gan
Shot Learning Decision Transformer Shot Adaptation Policy Generalization Offline Meta Reinforcement Learning

May 19, 2022

Concurrent Policy Blending and System Identification for Generalized Assistive Control
Luke Bhan, Marcos Quinones-Grueiro, Gautam Biswas
Collaborative Robot System Identification Policy Generalization Human Robot Collaborative Task Policy Blending