English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Accelerating artificial intelligence by photonics is an active field of studyaiming to exploit the unique properties of photons. Reinforcement learning isan important branch of machine learning, and photonic decision-makingprinciples have been demonstrated with respect to the multi-armed banditproblems. However, reinforcement learning could involve a massive number ofstates, unlike previously demonstrated bandit problems where the number ofstates is only one. Q-learning is a well-known approach in reinforcementlearning that can deal with many states. The architecture of Q-learning,however, does not fit well photonic implementations due to its separation ofupdate rule and the action selection. In this study, we organize a newarchitecture for multi-state reinforcement learning as a parallel array ofbandit problems in order to benefit from photonic decision-makers, which wecall parallel bandit architecture for reinforcement learning or PBRL in short.Taking a cart-pole balancing problem as an instance, we demonstrate that PBRLadapts to the environment in fewer time steps than Q-learning. Furthermore,PBRL yields faster adaptation when operated with a chaotic laser time seriesthan the case with uniformly distributed pseudorandom numbers where theautocorrelation inherent in the laser chaos provides a positive effect. We alsofind that the variety of states that the system undergoes during the learningphase exhibits completely different properties between PBRL and Q-learning. Theinsights obtained through the present study are also beneficial for existingcomputing platforms, not just photonic realizations, in acceleratingperformances by the PBRL algorithms and correlated random sequences.

Parallel bandit architecture based on laser chaos for reinforcement learning