Loading...

Brain Inspired Meta Reinforcement Learning Using Brain-Inspired Networks

Razavi Rohani, Roozbeh | 2022

308 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 55064 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Soleymani Baghshahi, Mahdih
  7. Abstract:
  8. Reinforcement learning is one of the most well-known learning paradigms in biological agents and one of the most used ones for solving plenty of problems. One of the reasons for this widespread use is the low demand for supervising signals. However, the sparsity of the reward signal causes increasing in sample complexity that needs for learning new tasks. This issue makes trouble in multi-task settings, specifically.One of the most promising approaches to learning new tasks by limited interaction with the environment is meta reinforcement learning. An approach in which fast adaption becomes possible by limiting hypothesis space and creating inductive biases by learning meta parameters. Recently, by observing the high efficiency of the proposed methods in this field, meta reinforcement learning has become one of the most important ways to get closer to general artificial intelligence. Therefore, in this research, with the usage of meta-learning in one way and with biological inspiration and specifically, cogitative neuroscience in the other way, trying to inject inductive biases that cause a reduction in sample complexity. In addition, we will test the efficiency of our model in a set of navigation problems. Meanwhile, we investigate the necessitate of existing all parts of the proposed model.
  9. Keywords:
  10. Deep Reinforcement Learning ; Exploration ; Memory ; Hierarchical Reinforcement Learning ; Cognitive Neuroscience ; Planning ; Predictive Representations

 Digital Object List