BarlowRL: Barlow Twins for Data-Efficient Reinforcement Learning
This addresses data efficiency for reinforcement learning practitioners, but it is incremental as it builds on existing methods like DER and Barlow Twins.
The paper tackled the problem of data inefficiency in reinforcement learning by introducing BarlowRL, which combines the Barlow Twins self-supervised learning framework with the DER algorithm, resulting in superior performance on the Atari 100k benchmark compared to DER and CURL.
This paper introduces BarlowRL, a data-efficient reinforcement learning agent that combines the Barlow Twins self-supervised learning framework with DER (Data-Efficient Rainbow) algorithm. BarlowRL outperforms both DER and its contrastive counterpart CURL on the Atari 100k benchmark. BarlowRL avoids dimensional collapse by enforcing information spread to the whole space. This helps RL algorithms to utilize uniformly spread state representation that eventually results in a remarkable performance. The integration of Barlow Twins with DER enhances data efficiency and achieves superior performance in the RL tasks. BarlowRL demonstrates the potential of incorporating self-supervised learning techniques to improve RL algorithms.