Results 41 to 50 of about 141,062 (268)
Knowledge distillation is one effective approach to compress deep learning models. However, the current distillation methods are relatively monotonous. There are still rare studies about the combination of distillation strategies using multiple types of ...
Ziyi Chen +5 more
doaj +1 more source
Knowledge Distillation with Adversarial Samples Supporting Decision Boundary
Many recent works on knowledge distillation have provided ways to transfer the knowledge of a trained network for improving the learning process of a new one, but finding a good technique for knowledge distillation is still an open problem. In this paper,
Choi, Jin Young +3 more
core +1 more source
MKD: Mixup-Based Knowledge Distillation for Mandarin End-to-End Speech Recognition
Large-scale automatic speech recognition model has achieved impressive performance. However, huge computational resources and massive amount of data are required to train an ASR model.
Xing Wu +4 more
doaj +1 more source
Residual Knowledge Distillation
9 pages, 3 figures, 3 ...
Gao, Mengya +3 more
openaire +2 more sources
NeuRes: Highly Activated Neurons Responses Transfer via Distilling Sparse Activation Maps
In recent years, Knowledge Distillation has obtained a significant interest in mobile, edge, and IoT devices due to its ability to transfer knowledge from the large and complex teacher to the lightweight student network.
Sharmen Akhter +3 more
doaj +1 more source
Progressive Label Distillation: Learning Input-Efficient Deep Neural Networks [PDF]
Much of the focus in the area of knowledge distillation has been on distilling knowledge from a larger teacher network to a smaller student network. However, there has been little research on how the concept of distillation can be leveraged to distill ...
Lin, Zhong Qiu, Wong, Alexander
core +2 more sources
Private Model Compression via Knowledge Distillation
The soaring demand for intelligent mobile applications calls for deploying powerful deep neural networks (DNNs) on mobile devices. However, the outstanding performance of DNNs notoriously relies on increasingly complex models, which in turn is associated
Bao, Weidong +5 more
core +1 more source
Dapagliflozin prevents methylglyoxal‐induced retinal cell death in ARPE‐19 cells
Diabetic macular oedema is a diabetes complication of the eye, which may lead to permanent blindness. ARPE‐19 are human retinal cells used to study retinal diseases and potential therapeutics. Methylglyoxal is a compound increased in uncontrolled diabetes due to elevated blood glucose.
Naina Trivedi +7 more
wiley +1 more source
A lightweight image classification method based on dual-source adaptive knowledge distillation
In the task of knowledge distillation, a dual-source adaptive knowledge distillation (DSAKD) method is proposed to address the issues of feature information loss during the feature alignment process and the lack of consideration for the differences in ...
ZHANG Kaibing, MA Dongtong, MENG Yalei
doaj +1 more source
Adversarially Robust Distillation
Knowledge distillation is effective for producing small, high-performance neural networks for classification, but these small networks are vulnerable to adversarial attacks.
Feizi, Soheil +3 more
core +1 more source

