We propose DoReFa-Net, a method to train convolutional neural networks that have low bitwidth weights and activations using low bitwidth parameter gradients. In particular, during backward pass, parameter gradients are stochastically quantized to low bitwidth numbers before being propagated to convolutional layers. As convolutions during forward/backward passes can now operate on low bitwidth weights and activations/gradients respectively, DoReFa-Net can use bit convolution kernels to accelerate both training and inference. Moreover, as bit convolutions can be efficiently implemented on CPU, FPGA, ASIC and GPU, DoReFa-Net opens the way to accelerate training of low bitwidth neural network on these hardware. Our experiments on SVHN and ImageNet datasets prove that DoReFa-Net can achieve comparable prediction accuracy as 32-bit counterparts. For example, a DoReFa-Net derived from AlexNet that has 1-bit weights, 2-bit activations, can be trained from scratch using 6-bit gradients to get 46.1% top-1 accuracy on ImageNet validation set. The DoReFa-Net AlexNet model is released publicly.

Quantized Neural Networks (QNNs), which use low bitwidth numbers for representing parameters and performing computations, have been proposed to reduce the computation complexity, storage size and memory usage. In QNNs, parameters and activations are uniformly quantized, such that the multiplications and additions can be accelerated by bitwise operations. However, distributions of parameters in Neural Networks are often imbalanced, such that the uniform quantization determined from extremal values may under utilize available bitwidth. In this paper, we propose a novel quantization method that can ensure the balance of distributions of quantized values. Our method first recursively partitions the parameters by percentiles into balanced bins, and then applies uniform quantization. We also introduce computationally cheaper approximations of percentiles to reduce the computation overhead introduced. Overall, our method improves the prediction accuracies of QNNs without introducing extra computation during inference, has negligible impact on training speed, and is applicable to both Convolutional Neural Networks and Recurrent Neural Networks. Experiments on standard datasets including ImageNet and Penn Treebank confirm the effectiveness of our method. On Ima-1 arXiv:1706.07145v1 [cs.CV] 22 Jun 2017 geNet, the top-5 error rate of our 4-bit quantized GoogLeNet model is 12.7%, which is superior to the state-of-the-arts of QNNs.

Cholesterol metabolism is often dysregulated in cancer. Squalene monooxygenase (SQLE) is the second rate-limiting enzyme involved in cholesterol synthesis. Since the discovery of SQLE dysregulation in cancer, compelling evidence has indicated that SQLE plays a vital role in cancer initiation and progression and is a promising therapeutic target for cancer treatment. In this review, we provide an overview of the role and regulation of SQLE in cancer and summarize the updates of antitumor therapy targeting SQLE.

An inappropriate diet is a risk factor for inflammatory bowel disease (IBD). It is established that the consumption of spicy food containing capsaicin is strongly associated with the recurrence and worsening of IBD symptoms. Moreover, capsaicin can induce neutrophil accumulation in the lamina propria, contributing to disease deterioration. To uncover the potential signaling pathway involved in capsaicin-induced relapse and the effects of capsaicin on neutrophil activation, we performed proteomic analyses of intestinal tissues from chronic colitis mice following capsaicin administration and transcriptomic analyses of dHL-60 cells after capsaicin stimulation. Collectively, these multiomic analyses identified proteins and genes that may be involved in disease flares, thereby providing new insights for future research.