Controlling chaotic itinerancy in laser dynamics for reinforcement learning

@article{Iwami2022ControllingCI,
  title={Controlling chaotic itinerancy in laser dynamics for reinforcement learning},
  author={Ryugo Iwami and Takatomo Mihana and Kazutaka Kanno and Satoshi Sunada and Makoto Naruse and Atsushi Uchida},
  journal={ArXiv},
  year={2022},
  volume={abs/2205.05987}
}
Photonic artificial intelligence has attracted considerable interest in accelerating machine learning; however, the unique optical properties have not been fully utilized for achieving higher-order functionalities. Chaotic itinerancy, with its spontaneous transient dynamics among multiple quasi-attractors, can be employed to realize brain-like functionalities. In this paper, we propose a method for controlling the chaotic itinerancy in a multi-mode semiconductor laser to solve a machine… 

Figures and Tables from this paper

Parallel bandit architecture based on laser chaos for reinforcement learning

This study organizes a new architecture for multi-state reinforcement learning as a parallel array of bandit problems in order to benefit from photonic decision-makers, which it is called parallel bandit architecture for reinforcement learning or PBRL in short.

Parallel photonic accelerator for decision making using optical spatiotemporal chaos

This work proposes a parallel photonic decision-making system for solving large-scale multi-armed bandit problems using optical spatiotemporal chaos, and solves a 512-arm bandit problem online, which is much larger than previous experiments by two orders of magnitude.

Chaotic mode-competition dynamics in a multimode semiconductor laser with optical feedback and injection

: Photonic computing is attracting increasing interest to accelerate information processing in machine learning applications. The mode-competition dynamics of multimode semiconductor lasers is useful

References

SHOWING 1-10 OF 59 REFERENCES

Scalable photonic reinforcement learning by time-division multiplexing of laser chaos

This study demonstrated a scalable, pipelined principle of resolving the multi-armed bandit problem by introducing time-division multiplexing of chaotically oscillated ultrafast time series to take advantage of the ultrahigh bandwidths of light wave and practical enabling technologies.

Ultrafast photonic reinforcement learning based on laser chaos

This work experimentally demonstrate that the ultrafast chaotic oscillatory dynamics of lasers efficiently solve the multi-armed bandit problem (MAB), which requires decision making concerning a class of difficult trade-offs called the exploration–exploitation dilemma.

Designing spontaneous behavioral switching via chaotic itinerancy

This study proposes a method for implementing chaotic itinerancy reproducibly in a high-dimensional chaotic neural network and demonstrates that this method enables us to easily design both the trajectories of quasi-attractors and the transition rules among them simply by adjusting the limited number of system parameters.

Arm order recognition in multi-armed bandit problem with laser chaos time series

This study applies to sectors where the order information is critical, such as efficient allocation of resources in information and communications technology, and has demonstrated numerically that this approach did improve arm order recognition accuracy significantly, along with reduced dependence on reward environments.

Laser network decision making by lag synchronization of chaos in a ring configuration.

A laser network consisting of unidirectionally coupled semiconductor lasers, whereby spontaneous exchange of the leader-laggard relationship in the lag synchronization of chaos is observed, is constructed, and a new direction in laser network-based decision making is suggested.

Decision making for the multi-armed bandit problem using lag synchronization of chaos in mutually coupled semiconductor lasers.

This study numerically and experimentally demonstrates the utilization of the synchronization of chaotic lasers for decision making, and successfully performs an on-line experimental demonstration of the decision making based on the two-laser coupled architecture.

On-chip photonic decision maker using spontaneous mode switching in a ring laser

This study paves the way for directly utilizing the fluctuating physics inherent in ring lasers, or integrated photonics technologies in general, for achieving or accelerating intelligent functionality.

Compact reservoir computing with a photonic integrated circuit.

This work experimentally demonstrates reservoir computing using a photonic integrated circuit with a semiconductor laser and a short external cavity, and proposes a method to increase the number of virtual nodes in delayed feedback using short node intervals and outputs from multiple delay times.

Chaotic itinerancy within the coupled dynamics between a physical body and neural oscillator networks

A series of simulations using different types of non-linear oscillator networks with a musculoskeletal model as a physical body to understand how the chaotic itinerancy of bodily behavior emerges from the coupled dynamics between the body and the brain.

Photonics for artificial intelligence and neuromorphic computing

Recent advances in integrated photonic neuromorphic neuromorphic systems are reviewed, current and future challenges are discussed, and the advances in science and technology needed to meet those challenges are outlined.
...