机器之心 & ArXiv Weekly Radiostation
参与:杜伟、楚航、罗若天
本周的重要论文包括谷歌大脑提出的拥有 1.6 万亿参数的语言模型 Switch Transformer,以及 IJCAI 2020 各奖项论文。
Read, Retrospect, Select: An MRC Framework to Short Text Entity Linking
Online 3D Bin Packing with Constrained Deep Reinforcement Learning
Synthesizing Aspect-Driven Recommendation Explanations from Reviews
A Multi-Objective Approach to Mitigate Negative Side Effects
SWITCH TRANSFORMERS: SCALING TO TRILLION PARAMETER MODELS WITH SIMPLE AND EFFICIENT SPARSITY
Deep learning-enabled medical computer vision
11 TOPS photonic convolutional accelerator for optical neural networks
ArXiv Weekly Radiostation:NLP、CV、ML 更多精选论文(附音频)
作者:Yingjie Gu、Xiaoye Qu、Zhefeng Wang 等
论文链接:https://arxiv.org/abs/2101.02394
作者:Hang Zhao、Qijin She、Chenyang Zhu 等
论文链接:https://arxiv.org/abs/2006.14978
作者:Trung-Hoang Le 和 Hady W. Lauw
论文链接:https://www.ijcai.org/Proceedings/2020/0336.pdf
作者:Sandhya Saisubramanian、Ece Kamar、Shlomo Zilberstein
论文链接:https://www.ijcai.org/Proceedings/2020/0050.pdf
作者:William Fedus、Barret Zoph、Noam Shazeer
论文链接:https://arxiv.org/pdf/2101.03961.pdf
作者:Andre Esteva、Katherine Chou、Serena Yeung 等
论文链接:https://www.nature.com/articles/s41746-020-00376-2#Sec6
作者:Xingyuan Xu、Mengxi Tan、Bill Corcoran 等
论文链接:https://www.nature.com/articles/s41586-020-03063-0