|
2017 ICLR 将于 4 月 26-27 日在法国东南部港口城市土伦举行。近日，大会接收论文名单公布。在 507 篇提交论文中，有 15 篇论文应邀进行演讲，poster 181 篇，应邀参加研讨会的有 48 篇。其中，FAIR 合著被接收的论文共 14 篇。 全部结果：https://openreview.net/forum?id=BkjLkSqxg
图片来自 Nando 推特 这 15 篇应邀演讲的论文分别是： Making Neural Programming Architectures Generalize via Recursion PDF:https://openreview.net/forum?id=BkbY4psgg Jonathon Cai, Richard Shin, Dawn Song End-to-end Optimized Image Compression PDFhttps://openreview.net/pdf?id=rJxdQ3jeg Johannes Ballé, Valero Laparra, Eero P. Simoncelli Optimization as a Model for Few-Shot Learning https://openreview.net/pdf?id=rJY0-Kcll Sachin Ravi, Hugo Larochelle Learning End-to-End Goal-Oriented Dialog https://openreview.net/pdf?id=S1Bb3D5gg Antoine Bordes, Y-Lan Boureau, Jason Weston Towards Principled Methods for Training Generative Adversarial Networks https://openreview.net/pdf?id=Hk4_qw5xe Martin Arjovsky, Leon Bottou Reinforcement Learning with Unsupervised Auxiliary Tasks https://openreview.net/pdf?id=SJ6yPD5xg Max Jaderberg, Volodymyr Mnih, Wojciech Marian Czarnecki, Tom Schaul, Joel Z Leibo, David Silver, Koray Kavukcuoglu Multi-Agent Cooperation and the Emergence of (Natural) Language https://openreview.net/forum?id=Hk8N3Sclg Angeliki Lazaridou, Alexander Peysakhovich, Marco Baroni Understanding deep learning requires rethinking generalization https://openreview.net/forum?id=Sy8gdB9xx Chiyuan Zhang, Samy Bengio, Moritz Hardt, Benjamin Recht, Oriol Vinyals Neural Architecture Search with Reinforcement Learning https://openreview.net/pdf?id=r1Ue8Hcxg Barret Zoph, Quoc Le Q-Prop: Sample-Efficient Policy Gradient with An Off-Policy Critic https://openreview.net/forum?id=SJ3rcZcxl Shixiang Gu, Timothy Lillicrap, Zoubin Ghahramani, Richard E. Turner, Sergey Levine Learning to Act by Predicting the Future https://openreview.net/forum?id=rJLS7qKe Alexey Dosovitskiy, Vladlen Koltun On Large-Batch Training for Deep Learning: Generalization Gap and Sharp Minima https://openreview.net/forum?id=H1oyRlYgg Nitish Shirish Keskar, Dheevatsa Mudigere, Jorge Nocedal, Mikhail Smelyanskiy, Ping Tak Peter Tang Semi-supervised Knowledge Transfer for Deep Learning from Private Training Data https://openreview.net/pdf?id=HkwoSDPgg Nicolas Papernot, Martín Abadi, Úlfar Erlingsson, Ian Goodfellow, Kunal Talwar Amortised MAP Inference for Image Super-resolution https://openreview.net/forum?id=S1RP6GLle Casper Kaae Sønderby, Jose Caballero, Lucas Theis, Wenzhe Shi, Ferenc Huszár Learning Graphical State Transitions https://openreview.net/forum?id=HJ0NvFzxl Daniel D. Johnson 另外，Nando 在推特上公开表示了对以下几篇论文的赞赏和喜爱。 论文：Third Person Imitation Learning， 作者： Bradly C Stadie, Pieter Abbeel, Ilya Sutskever
论文：Modular Multitask Reinforcement Learning with Policy Sketches 作者：Jacob Andreas, Dan Klein, Sergey Levine
论文：Optimization as a Model for Few-Shot Learning 作者：Sachin Ravi, Hugo Larochelle
论文：Making Neural Programming Architectures Generalize via Recursion（被 Nando 形容为突破） 作者：Jonathon Cai, Richard Shin, Dawn Song
不过，有趣的是，2016 年刷爆各路媒体的 LipNet 论文却出人意料地遭拒。 (责任编辑：本港台直播) |
