Policy Network

Policy networks are neural networks used in reinforcement learning to map states to actions, aiming to learn optimal control policies for agents in various environments. Current research focuses on improving efficiency and generalization of these networks, exploring architectures like diffusion models and leveraging techniques such as meta-learning, constrained optimization, and Lipschitz-bounded parameterizations to enhance robustness and sample efficiency. This work is significant for advancing artificial intelligence, particularly in robotics and autonomous systems, by enabling the development of more adaptable, reliable, and efficient agents capable of handling complex tasks and uncertain environments.

Papers

October 17, 2022

A Novel Membership Inference Attack against Dynamic Neural Networks by Utilizing Policy Networks Information
Pan Li, Peizhuo Lv, Shenchen Zhu, Ruigang Liang, Kai Chen
Membership Inference Attack Dynamic Neural Network Policy Network Static Deep Neural Network

September 5, 2022

A New Approach to Training Multiple Cooperative Agents for Autonomous Driving
Ruiyang Yang, Siheng Li, Beihong Jin
Autonomous Driving Novel Approach Multiple Agent Multi Agent Cooperation Policy Network Cooperative Control

July 7, 2022

Hyper-Universal Policy Approximation: Learning to Generate Actions from a Single Image using Hypernets
Dimitrios C. Gklezakos, Rishi Jha, Rajesh P. N. Rao
Affordance Learning Single Image Action Generation Policy Network Conditioned Policy Expressive Policy

July 4, 2022

General Policy Evaluation and Improvement by Learning to Identify Few But Crucial States
Francesco Faccio, Aditya Ramesh, Vincent Herrmann, Jean Harb, Jürgen Schmidhuber
Reinforcement Learning Policy Evaluation Large Relevance Improvement Near Optimal Policy Policy Network Linear Policy

April 15, 2022

Understanding Game-Playing Agents with Natural Language Annotations
Nicholas Tomlin, Andre He, Dan Klein
Policy Network Game Playing Agent Game Model Natural Language Annotation

February 28, 2022

GPU-Accelerated Policy Optimization via Batch Automatic Differentiation of Gaussian Processes for Real-World Control
Abdolreza Taheri, Joni Pajarinen, Reza Ghabcheloo
Gaussian Process Policy OpTimization Policy Optimization Automatic Differentiation Policy Network Real World Control

February 11, 2022

Fast Model-based Policy Search for Universal Policy Networks
Buddhika Laknath Semage, Thommen George Karimpanal, Santu Rana, Svetha Venkatesh
Reinforcement Learning Gaussian Process Policy Network Policy Search Model Based Policy

November 15, 2021

Modular Networks Prevent Catastrophic Interference in Model-Based Multi-Task Reinforcement Learning
Robin Schiewer, Laurenz Wiskott
Multi Task Reinforcement Learning Learned Dynamic Policy Network Modular Network Multi Task Transfer

Policy Network

Papers

A Novel Membership Inference Attack against Dynamic Neural Networks by Utilizing Policy Networks Information

A New Approach to Training Multiple Cooperative Agents for Autonomous Driving

Hyper-Universal Policy Approximation: Learning to Generate Actions from a Single Image using Hypernets

General Policy Evaluation and Improvement by Learning to Identify Few But Crucial States

Understanding Game-Playing Agents with Natural Language Annotations

GPU-Accelerated Policy Optimization via Batch Automatic Differentiation of Gaussian Processes for Real-World Control

Fast Model-based Policy Search for Universal Policy Networks

Modular Networks Prevent Catastrophic Interference in Model-Based Multi-Task Reinforcement Learning