Search Constraints

Start Over
Filtering by: Creator Kuhta, Nicholas A. Remove constraint Creator: Kuhta, Nicholas A. Contributor Agostinelli, Victor Remove constraint Contributor: Agostinelli, Victor Degree Field Computer Science Remove constraint Degree Field: Computer Science

Search Results

  • Partitioning-Guided K-Means for Effective Empty Cluster Resolution in Extreme Model Compression

    Creator:
    Huang, Tianhong
    Abstract:
    Compactness in deep learning can be critical to a model’s viability in low-resource applications, and a common approach to extreme model compression is quantization. We consider Iterative Product Quantization (iPQ) with Quant-Noise [Fan et al., 2020] to be state-of-the-art in this area, but this quantization framework suffers from preventable inference...
    Resource Type:
    Masters Thesis
    Full Text:
    ://ideas.repec.org/a/spr/jclass/v34y2017i2d10.1007\ _s00357-017-9231-0.html. Yoshua Bengio, Nicholas Léonard, and