Fast and Data Efficient Reinforcement Learning from Pixels via Non-Parametric Value Approximation

Add code
Mar 07, 2022
Alexander Long, Alan Blair, Herke van Hoof

Figure 1 for Fast and Data Efficient Reinforcement Learning from Pixels via Non-Parametric Value Approximation
Figure 2 for Fast and Data Efficient Reinforcement Learning from Pixels via Non-Parametric Value Approximation
Figure 3 for Fast and Data Efficient Reinforcement Learning from Pixels via Non-Parametric Value Approximation
Figure 4 for Fast and Data Efficient Reinforcement Learning from Pixels via Non-Parametric Value Approximation

Share this with someone who'll enjoy it:

Twitter IconFacebook IconLinkedin IconWhatsapp IconMessenger Icon

We present Nonparametric Approximation of Inter-Trace returns (NAIT), a Reinforcement Learning algorithm for discrete action, pixel-based environments that is both highly sample and computation efficient. NAIT is a lazy-learning approach with an update that is equivalent to episodic Monte-Carlo on episode completion, but that allows the stable incorporation of rewards while an episode is ongoing. We make use of a fixed domain-agnostic representation, simple distance based exploration and a proximity graph-based lookup to facilitate extremely fast execution. We empirically evaluate NAIT on both the 26 and 57 game variants of ATARI100k where, despite its simplicity, it achieves competitive performance in the online setting with greater than 100x speedup in wall-time.

* AAAI2022  
View paper onarxiv icon

Share this with someone who'll enjoy it:

Twitter IconFacebook IconLinkedin IconWhatsapp IconMessenger Icon