Strategies for Affect-Controlled Action-Selection in Soar-RL (2007)
Proceedings of the 2nd. International Work-Conference on the Interplay between Natural and Artificial Computation (IWINAC)
Reinforcement learning (RL) agents can benefit from adaptive exploration/exploitation behavior, especially in dynamic environments. We focus on regulating this exploration/exploitation behavior by controlling the action-selection mechanism of RL. Inspired by psychological studies which show that affect influences human decision making, we use artificial affect to influence an agent’s action-selection. Two existing affective strategies are implemented and, in addition, a new hybrid method that combines both. These strategies are tested on ‘maze tasks’ in which a RL agent has to find food (rewarded location) in a maze. We use Soar-RL, the new RL-enabled version of Soar, as a model environment. One task tests the ability to quickly adapt to an environmental change, while the other tests the ability to escape a local optimum in order to find the global optimum. We show that artificial affect-controlled action-selection in some cases helps agents to faster adapt to changes in the environment. [via]
http://www.liacs.nl/~broekens/files/Hogewonin...
strategy
Related Files
- The Importance of Action History in Decision Making and Reinforcement Learning
- Soar-RL: Integrating Reinforcement Learning with Soar
- On the Performance of Internet Worm Scanning Strategies
- Experiments in Transfer Across Multiple Learning Mechanisms
- Comprehensive Working Memory Activation in Soar
- Characteristics of Long-term Learning in Soar and its Application to the Utility Problem
Sponsored Links
Free Download Harman/Kardon Manual, Guide, Instructions, available in PDF ebooks format.
Rate this Document
Popular eBooks
- Java Platform Migration Guide Version 1.3 to 5.0
- Java Web Start Overview White Paper
- Java 2 Platform, Standard Edition 5.0: Troubleshooting and Diagnostic Guide
- MySQL 5.0 Stored Procedures
- Numerical Python
- An Introduction to Python - Release 2.2.2
- Python/C API Reference Manual - Release 2.5
- Trusted Platform Module (TPM) based Security on Notebook PCs - White Paper
- A Z Patterns Catalogue II: Definitions and Laws, v0.1
- Tutorial on Python Iterators and Generators
ADS
Tag Clouds
3d algorithm analysis application architecture design ebook energy framework guide java javascript kernel linux manual network paper programming python reference security software system theory thesis tutorial typography web service white paper xml
Last Download
- The Elements of Typographic Style
- Automatic Kinetic Typography Composer
- PyFemax : A Python Finite Element Maxwell Solver
- Adobe Premiere Pro CS4 4.1
- Coot Tutorial II: More Advanced Usage
BookShelf
- Linux Security Quick Reference Guide
- Using A Genetic Algorithm to Identify Predictive Bayesian Models in Medical Informatics
- Anti-Disassembly using Cryptographic Hash Functions
- Research on Aerodynamic Drag Reduction by Vortex Generators
- Women Carrying Water: At the Crossroads of Technology and Critical Theory
