17 citations · 39 across the 6 of their papers we have counts for
7 papers
Cross-Modal Knowledge Transfer Without Task-Relevant Source Data
Sk Miraj Ahmed, Suhas Lohit, Kuan-Chuan Peng +2
Cost-effective depth and infrared sensors as alternatives to usual RGB sensors are now a reality, and have some advantages over RGB in domains like autonomous navigation and remote…
Combining Data-driven Supervision with Human-in-the-loop Feedback for Entity Resolution
Wenpeng Yin, Shelby Heinecke, Jia Li +7
The distribution gap between training datasets and data encountered in production is well acknowledged. Training datasets are often constructed over a fixed period of time and by c…
To Boost or not to Boost: On the Limits of Boosted Neural Networks
Sai Saketh Rambhatla, Michael Jones, Rama Chellappa
Boosting is a method for finding a highly accurate hypothesis by linearly combining many ``weak" hypotheses, each of which may be only moderately accurate. Thus, boosting is a meth…
Model Compression Using Optimal Transport
Suhas Lohit, Michael Jones
Model compression methods are important to allow for easier deployment of deep learning models in compute, memory and energy-constrained environments such as mobile phones. Knowled…
Multi-head Knowledge Distillation for Model Compression
Huan Wang, Suhas Lohit, Michael Jones +1
Several methods of knowledge distillation have been developed for neural network compression. While they all use the KL divergence loss to align the soft outputs of the student mod…
What Makes a "Good" Data Augmentation in Knowledge Distillation -- A Statistical Perspective
Huan Wang, Suhas Lohit, Mike Jones +1
Knowledge distillation (KD) is a general neural network training approach that uses a teacher model to guide the student model. Existing works mainly study KD from the network outp…