1. Dağıtılmış sistemler için tekrarlanan sinir ağları merkezli çevrimiçi öğrenim algoritmaları
- Author
-
Suleyman S. Kozat, Tolga Ergen, and S. Onur Sahin
- Subjects
Structure (mathematical logic) ,State-space representation ,Long short term memory networks ,business.industry ,Computer science ,Online learning ,Node (networking) ,020206 networking & telecommunications ,02 engineering and technology ,Sequential regression ,Distributed systems ,Online training ,Set (abstract data type) ,Recurrent neural network ,0202 electrical engineering, electronic engineering, information engineering ,020201 artificial intelligence & image processing ,Artificial intelligence ,Particle filter ,business - Abstract
Date of Conference: 2-5 May 2018 In this paper, we investigate online parameter learning for Long Short Term Memory (LSTM) architectures in distributed networks. Here, we first introduce an LSTM based structure for regression. Then, we provide the equations of this structure in a state space form for each node in our network. Using this form, we then learn the parameters via our Distributed Particle Filtering based (DPF) training method. Our training method asymptotically converges to the optimal parameter set provided that we satisfy certain trivial requirements. While achieving this performance, our training method only causes a computational load that is similar to the efficient first order gradient based training methods. Through real life experiments, we show substantial performance gains compared to the conventional methods.
- Published
- 2018