Reinforcement Learning Architecture

Reinforcement learning architecture focuses on designing efficient and effective structures for training agents to make optimal decisions in complex environments. Current research emphasizes improving sample efficiency through techniques like hierarchical policies, attention mechanisms, and the incorporation of uncertainty representations (e.g., Kalman filters) into models, particularly for partially observable scenarios. These advancements are crucial for enabling robust performance in real-world applications, such as robotics, autonomous driving, and resource management in cloud computing, where data is often limited or noisy. Furthermore, research explores incorporating ethical considerations and reasoning into agent design, aiming to create more responsible and predictable AI systems.

Papers

September 25, 2024

Uncertainty Representations in State-Space Layers for Deep Reinforcement Learning under Partial Observability
Carlos E. Luis, Alessandro G. Bottero, Julia Vinogradska, Felix Berkenkamp, Jan Peters
Deep Reinforcement Learning Kalman Filter Partial Observability Reinforcement Learning Architecture Uncertainty Representation State Space Layer

September 23, 2024

Acting for the Right Reasons: Creating Reason-Sensitive Artificial Moral Agents
Kevin Baum, Lisa Dargasz, Felix Jahn, Timo P. Gros, Verena Wolf
Moral Judgment Moral Agent Reinforcement Learning Architecture Moral Decision Shield Synthesis

September 20, 2024

From Cognition to Precognition: A Future-Aware Framework for Social Navigation
Zeying Gong, Tianshuai Hu, Ronghe Qiu, Junwei Liang
Social Navigation Cognitive Learning Human Trajectory Reinforcement Learning Architecture Long Term Motion

July 27, 2024

On the benefits of pixel-based hierarchical policies for task generalization
Tudor Cristea-Platon, Bogdan Mazoure, Josh Susskind, Walter Talbott
Complementary Benefit Task Generalization Hierarchical Policy Reinforcement Learning Architecture

May 22, 2024

Learning To Play Atari Games Using Dueling Q-Learning and Hebbian Plasticity
Md Ashfaq Salehin
Deep Q Network Neural Agent ATARI Game Hebbian Plasticity Reinforcement Learning Architecture Model Plasticity

April 9, 2024

Emergent Braitenberg-style Behaviours for Navigating the ViZDoom `My Way Home' Labyrinth
Caleidgh Bayer, Robert J. Smith, Malcolm I. Heywood
Simple Heuristic Heuristic Search Joint Entity Reinforcement Learning Architecture

August 11, 2023

A Deep Recurrent-Reinforcement Learning Method for Intelligent AutoScaling of Serverless Functions
Siddharth Agarwal, Maria A. Rodriguez, Rajkumar Buyya
Proximal Policy Optimization Model Free Serverless Computing Reinforcement Learning Architecture Auto Scaling

February 24, 2023

Model-Based Uncertainty in Value Functions
Carlos E. Luis, Alessandro G. Bottero, Julia Vinogradska, Felix Berkenkamp, Jan Peters
Model Based Reinforcement Learning Value Function Uncertainty Modeling Bellman Equation Reinforcement Learning Architecture Inefficient Exploration

August 22, 2022

Event-Triggered Model Predictive Control with Deep Reinforcement Learning for Autonomous Driving
Fengying Dang, Dong Chen, Jun Chen, Zhaojian Li
Deep Reinforcement Learning Autonomous Driving Model Predictive Control Reinforcement Learning Architecture

April 21, 2022

Learning how to Interact with a Complex Interface using Hierarchical Reinforcement Learning
Gheorghe Comanici, Amelia Glaese, Anita Gergely, Daniel Toyama, Zafarali Ahmed, Tyler Jackson, Philippe Hamel, Doina Precup
Hierarchical Reinforcement Learning DQN Agent User Interface Interactive Agent Reinforcement Learning Architecture

February 1, 2022

Improving Sample Efficiency of Value Based Models Using Attention and Vision Transformers
Amir Ardalan Kalantari, Mohammad Amini, Sarath Chandar, Doina Precup
Vision Transformer Human Attention Neural Architecture Sample Efficiency Valuation Model Reinforcement Learning Architecture

December 13, 2021

Human-Level Control through Directly-Trained Deep Spiking Q-Networks
Guisong Liu, Wenjie Deng, Xiurui Xie, Li Huang, Huajin Tang
Neuromorphic Hardware Deep Q Network Reinforcement Learning Architecture Human Control

Reinforcement Learning Architecture

Papers

Uncertainty Representations in State-Space Layers for Deep Reinforcement Learning under Partial Observability

Acting for the Right Reasons: Creating Reason-Sensitive Artificial Moral Agents

From Cognition to Precognition: A Future-Aware Framework for Social Navigation

On the benefits of pixel-based hierarchical policies for task generalization

Learning To Play Atari Games Using Dueling Q-Learning and Hebbian Plasticity

Emergent Braitenberg-style Behaviours for Navigating the ViZDoom `My Way Home' Labyrinth

A Deep Recurrent-Reinforcement Learning Method for Intelligent AutoScaling of Serverless Functions

Model-Based Uncertainty in Value Functions

Event-Triggered Model Predictive Control with Deep Reinforcement Learning for Autonomous Driving

Learning how to Interact with a Complex Interface using Hierarchical Reinforcement Learning

Improving Sample Efficiency of Value Based Models Using Attention and Vision Transformers

Human-Level Control through Directly-Trained Deep Spiking Q-Networks