Search Results for author:
Found 5 papers, 2 papers with code
Improve Knowledge Distillation via Label Revision and Data Selection
In addition to the supervision of ground truth, the vanilla KD method regards the predictions of the teacher as soft labels to supervise the training of the student model.
Feature Fusion from Head to Tail for Long-Tailed Visual Recognition
To rectify this issue, we propose to augment tail classes by grafting the diverse semantic information from head classes, referred to as head-to-tail fusion (H2T).
Adjusting Logit in Gaussian Form for Long-Tailed Visual Recognition
Based on these perturbed features, two novel logit adjustment methods are proposed to improve model performance at a modest computational overhead.
Compact Neural Networks via Stacking Designed Basic Units
To this end, this paper presents a new method termed TissueNet, which directly constructs compact neural networks with fewer weight parameters by independently stacking designed basic units, without requiring additional judgement criteria anymore.
Compressing Deep Convolutional Neural Networks by Stacking Low-dimensional Binary Convolution Filters
One popular way to reduce the memory cost of deep CNN model is to train binary CNN where the weights in convolution filters are either 1 or -1 and therefore each weight can be efficiently stored using a single bit.
