no code implementations • 13 Apr 2023 • Deng Pan, Mohammad Ali Khoshkholghi, Toktam Mahmoodi
Those two methods can effectively control energy consumption and communication cost by controlling the number of local training epochs, local communication, and global communication.
1 code implementation • 17 Jan 2023 • Xin Li, Deng Pan, Chengyin Li, Yao Qiang, Dongxiao Zhu
There are increasing demands for understanding deep neural networks' (DNNs) behavior spurred by growing security and/or transparency concerns.
no code implementations • 23 Nov 2022 • Xin Li, Xiangrui Li, Deng Pan, Yao Qiang, Dongxiao Zhu
Deep neural networks (DNNs) for supervised learning can be viewed as a pipeline of the feature extractor (i. e., last hidden layer) and a linear classifier (i. e., output layer) that are trained jointly with stochastic gradient descent (SGD) on the loss function (e. g., cross-entropy).
1 code implementation • 14 Dec 2020 • Xin Li, Xiangrui Li, Deng Pan, Dongxiao Zhu
This inspires us to propose a new Probabilistically Compact (PC) loss with logit constraints which can be used as a drop-in replacement for cross-entropy (CE) loss to improve CNN's adversarial robustness.
no code implementations • 12 Jul 2020 • Deng Pan, Xiangrui Li, Xin Li, Dongxiao Zhu
Latent factor collaborative filtering (CF) has been a widely used technique for recommender system by learning the semantic representations of users and items.
1 code implementation • 24 Jun 2020 • Xin Li, Deng Pan, Dongxiao Zhu
Medical imaging AI systems such as disease classification and segmentation are increasingly inspired and transformed from computer vision based AI systems.
1 code implementation • 4 Mar 2020 • Xiangrui Li, Xin Li, Deng Pan, Dongxiao Zhu
Deep convolutional neural networks (CNNs) trained with logistic and softmax losses have made significant advancement in visual recognition tasks in computer vision.
no code implementations • 23 Feb 2020 • Xiangrui Li, Deng Pan, Xin Li, Dongxiao Zhu
In each iteration of SGD, a mini-batch from the training data is sampled and the true gradient of the loss function is estimated as the noisy gradient calculated on this mini-batch.