Summary of Sup3r: a Semi-supervised Algorithm For Increasing Sparsity, Stability, and Separability in Hierarchy Of Time-surfaces Architectures, by Marco Rasetto et al.
Sup3r: A Semi-Supervised Algorithm for increasing Sparsity, Stability, and Separability in Hierarchy Of Time-Surfaces architectures
by Marco Rasetto, Himanshu Akolkar, Ryad Benosman
First submitted to arxiv on: 15 Apr 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Neural and Evolutionary Computing (cs.NE)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary Sup3r, a Semi-Supervised algorithm, tackles challenges in accuracy and compatibility with neuromorphic hardware faced by the Hierarchy Of Time-Surfaces (HOTS) algorithm. Sup3r enhances sparsity, stability, and separability in HOTS networks, replacing external classifiers through end-to-end online training using semi-supervised learning. It learns class-informative patterns, mitigates confounding features, and reduces processed events. Sup3r also facilitates continual and incremental learning, allowing adaptation to data distribution shifts and learning new tasks without forgetting. Preliminary results on N-MNIST demonstrate comparable accuracy to similarly sized Artificial Neural Networks trained with back-propagation. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Sup3r is a new way to make HOTS networks better. It helps solve problems that make it hard for HOTS to work well with special hardware. Sup3r makes the network more efficient, learns patterns that help separate things, and gets rid of extra information. It also lets the network keep learning even when it’s given new data or has to learn a new task without forgetting what it already knows. In tests on N-MNIST, Sup3r did almost as well as other networks trained in a different way. |
Keywords
» Artificial intelligence » Semi supervised
