Home

Published

- 2 min read

A Unified Framework for Neural Computation and Learning Over Time

NeuralNetworks Foundations
img of A Unified Framework for Neural Computation and Learning Over Time

Report

Details

Hamiltonian Learning is a novel learning framework designed to handle problems of learning over time, insipired by Optimal Control Theory. This paper showcases the links with the largely known gradient-based optimization and future perspectives.

  • Authors: Stefano Melacci, Alessandro Betti, Michele Casoni, Tommaso Guidi, Matteo Tiezzi, Marco Gori
  • Title: **A Unified Framework for Neural Computation and Learning Over Time
  • Where: Technical Report (arXiv) - under review

BibTeX

   @misc{melacci2024unifiedframeworkneuralcomputation,
      title={A Unified Framework for Neural Computation and Learning Over Time}, 
      author={Stefano Melacci and Alessandro Betti and Michele Casoni and Tommaso Guidi and Matteo Tiezzi and Marco Gori},
      year={2024},
      eprint={2409.12038},
      archivePrefix={arXiv},
      primaryClass={cs.LG},
      url={https://arxiv.org/abs/2409.12038}, 
}