Aidin Attar PhD student · Information Engineering

Research

My work focuses on spiking neural networks and learning algorithms that use local information. I am interested in how these models can learn over time and operate under the constraints of neuromorphic hardware.

Main topics

Spiking Neural Networks

  • Deep and structured SNN architectures
  • Temporal coding and event-based processing
  • STDP and reward-modulated plasticity

Local and Continual Learning

  • Learning without global backpropagation
  • Fast adaptation and slower memory consolidation
  • Online learning and stability under changing conditions

Neuromorphic Computing

  • Hardware-aware model design
  • Sparse, event-driven computation
  • Learning under memory and energy constraints

My background in physics also informs my interest in dynamical systems, predictive processing, and neural models inspired by neuroscience. I develop experimental software to investigate these ideas and evaluate them on spiking and event-based datasets.