7 papers | PyTorch官方框架论文;浙大阿里等新方法提升唇读效果
本周既有深度学习社区非常关注的 PyTorch 框架论文,还有浙大和阿里巴巴等利用学习语音识别器增强唇读效果的研究。
PyTorch: An Imperative Style, High-Performance Deep Learning Library
On The Equivalence Between Node Embeddings And Structural Graph Representations
SEAN: Image Synthesis with Semantic Region-Adaptive Normalization
Hearing Lips: Improving Lip Reading by Distilling Speech Recognizers
Neural Academic Paper Generation
DEEP DOUBLE DESCENT: WHERE BIGGER MODELS AND MORE DATA HURT
A Simple Proof of the Quadratic Formula
论文链接:https://papers.nips.cc/paper/9015-pytorch-an-imperative-style-high-performance-deep-learning-library.pdf
论文链接:https://arxiv.org/pdf/1910.00452.pdf
论文链接:https://arxiv.org/abs/1911.12861
论文链接:https://arxiv.org/pdf/1911.11502.pdf
论文链接:https://arxiv.org/abs/1912.01982
论文链接:https://mltheory.org/deep.pdf
论文链接:https://arxiv.org/pdf/1910.06709.pdf