Arxiv paper – MCNC: MANIFOLD-CONSTRAINED REPARAMETERIZATION FOR NEURAL COMPRESSION

In this episode, we discuss MCNC: MANIFOLD-CONSTRAINED REPARAMETERIZATION FOR NEURAL COMPRESSION by The authors of the paper are: – Chayne Thrash – Ali Abbasi – Reed Andreas – Parsa Nooralinejad – Soroush Abbasi Koohpayegani – Hamed Pirsiavash – Soheil Kolouri. The paper introduces Manifold-Constrained Neural Compression (MCNC), a novel model compression technique that confines parameters to low-dimensional, pre-defined nonlinear manifolds. This approach leverages the over-parameterization of deep networks to find high-quality solutions while achieving superior compression rates. Experiments across computer vision and NLP tasks show that MCNC outperforms existing methods in compression efficiency, accuracy, and reconstruction speed.