Uncovering the Origins of Instability in Dynamical Systems: How Attention Mechanism Can Help?

no code implementations • 19 Dec 2022 • Nooshin Bahador, Milad Lankarany

The results demonstrated that the nodes receiving more attention cause more instability in the system.

Deep Direct Discriminative Decoders for High-dimensional Time-series Data Analysis

2 code implementations • 22 May 2022 • Mohammad R. Rezaei, Milos R. Popovic, Milad Lankarany, Ali Yousefi

The D4 brings deep neural networks' expressiveness and scalability to the SSM formulation letting us build a novel solution that efficiently estimates the underlying state processes through high-dimensional observation signal.

Hippocampus Time Series +2

Gradient-Free Neural Network Training via Synaptic-Level Reinforcement Learning

no code implementations • 29 May 2021 • Aman Bhargava, Mohammad R. Rezaei, Milad Lankarany

Trained MLPs yield character recognition performance comparable to identically shaped networks trained with gradient descent.

reinforcement-learning Reinforcement Learning (RL)