Reproduce “LLC: Accurate, Multi-purpose Learnt Low-dimensional Binary Codes” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Aditya Kusupati, Matthew Wallingford, Vivek Ramanujan, Raghav Somani, Jae Sung Park, Krishna Pillutla, Prateek Jain, Sham Kakade and 1 more
LLC learns short binary codes for image classes and images from labeled classification data, using class codes to supervise image codes without extra attributes or metadata. On ImageNet-1K, 20-bit codes reach 74.57% top-1 accuracy with minimum Hamming distance decoding, versus 77% for real-valued features; retrieval results also compare favorably with HashNet, but end-to-end speedups are not measured.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.