Compressive Classification (Machine Learning without learning)
arXiv:1812.01410v17 citations
Originality Synthesis-oriented
AI Analysis
This work addresses the challenge of efficient classification for machine learning practitioners by enabling learning from compressed data, though it appears incremental as it extends an existing framework to a new task.
The paper tackles the problem of classification within the compressive learning framework by proposing a new method that uses compressed data summaries instead of full datasets, and introduces a novel sketch function specifically for images.
Compressive learning is a framework where (so far unsupervised) learning tasks use not the entire dataset but a compressed summary (sketch) of it. We propose a compressive learning classification method, and a novel sketch function for images.