On Neural Consolidation for Transfer in Reinforcement Learning

10/05/2022
by   Valentin Guillet, et al.
0

Although transfer learning is considered to be a milestone in deep reinforcement learning, the mechanisms behind it are still poorly understood. In particular, predicting if knowledge can be transferred between two given tasks is still an unresolved problem. In this work, we explore the use of network distillation as a feature extraction method to better understand the context in which transfer can occur. Notably, we show that distillation does not prevent knowledge transfer, including when transferring from multiple tasks to a new one, and we compare these results with transfer without prior distillation. We focus our work on the Atari benchmark due to the variability between different games, but also to their similarities in terms of visual features.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
12/29/2019

Real-time Policy Distillation in Deep Reinforcement Learning

Policy distillation in deep reinforcement learning provides an effective...
research
10/05/2022

Neural Distillation as a State Representation Bottleneck in Reinforcement Learning

Learning a good state representation is a critical skill when dealing wi...
research
02/06/2019

Distilling Policy Distillation

The transfer of knowledge from one policy to another is an important too...
research
10/30/2018

Exploration by Random Network Distillation

We introduce an exploration bonus for deep reinforcement learning method...
research
02/06/2020

Transfer Heterogeneous Knowledge Among Peer-to-Peer Teammates: A Model Distillation Approach

Peer-to-peer knowledge transfer in distributed environments has emerged ...
research
05/31/2021

Procedural Content Generation: Better Benchmarks for Transfer Reinforcement Learning

The idea of transfer in reinforcement learning (TRL) is intriguing: bein...
research
03/01/2018

Knowledge Transfer with Jacobian Matching

Classical distillation methods transfer representations from a "teacher"...

Please sign up or login with your details

Forgot password? Click here to reset