Current browse context:
cs.LG
Change to browse by:
References & Citations
Condensed Matter > Soft Condensed Matter
Title: Desynchronous Learning in a Physics-Driven Learning Network
(Submitted on 10 Jan 2022 (v1), last revised 1 Dec 2022 (this version, v2))
Abstract: In a neuron network, synapses update individually using local information, allowing for entirely decentralized learning. In contrast, elements in an artificial neural network (ANN) are typically updated simultaneously using a central processor. Here we investigate the feasibility and effect of desynchronous learning in a recently introduced decentralized, physics-driven learning network. We show that desynchronizing the learning process does not degrade performance for a variety of tasks in an idealized simulation. In experiment, desynchronization actually improves performance by allowing the system to better explore the discretized state space of solutions. We draw an analogy between desynchronization and mini-batching in stochastic gradient descent, and show that they have similar effects on the learning process. Desynchronizing the learning process establishes physics-driven learning networks as truly fully distributed learning machines, promoting better performance and scalability in deployment.
Submission history
From: Sam Dillavou [view email][v1] Mon, 10 Jan 2022 05:38:01 GMT (4663kb,D)
[v2] Thu, 1 Dec 2022 21:36:04 GMT (4668kb,D)
Link back to: arXiv, form interface, contact.