长按二维码,关注官方公众号
使用微信扫一扫分享到朋友圈
使用微信扫一扫进入小程序分享活动
作为当前全球最负盛名的 AI 学术会议之一，NeurIPS 是每年AI界的重要事件，通常在 12 月举办。大会讨论的内容包含深度学习、计算机视觉、大规模机器学习、学习理论、优化、稀疏理论等众多细分领域。今年 NeurIPS 已是第 36 届，将于 11 月 28 日至 12 月 9 日举行，为期两周。第一周将在美国新奥尔良 Ernest N. Morial 会议中心举行现场会议，第二周改为线上会议。9 月下旬，NeurIPS 公布了今年的论文接收情况，一共提交了 10411 篇论文，2672 篇获接收，最终接收率为 25.6%。
为了给国内 AI 社区从业者搭建一个自由轻松的学术交流平台，机器之心将于 2022 年 11 月 26 日线上举办「2022 NeurIPS Meetup China」学术交流活动，广邀 AI 社区成员参与。今日，「2022 NeurIPS Meetup China」全日程正式公布，活动包含 4 个 Keynote、11 篇论文分享与多场企业招聘环节。
开场
大规模多模态预训练的最新研究进展
机器学习驱动的求解器研究
Improving 3D-aware Image Synthesis with A Geometry Dis criminator
PIE-G：Pretrained Image Encoder for Generalizable Visual Reinforcement Learning
Rethinking Lipschitz Neural Networks and Certified Robustness: A Boolean Function Perspective
Learning Neural Set Functions Under the Optimal Subset Oracle
TalentAI企业招聘
Towards Open-world Reinforcement Learning
数据高效的深度视觉识别
Jump Self-attention: Capturing High-order Statistics in Transformers
Multiagent Q-learning with Sub-Team Coordination
Convolutional Nerual Networks on Graphs with Chebyshev Approximation, Revisited
A Policy-Guided Imitation Approach for Offline Reinforcement Learning
Uncovering the Structural Fairness in Graph Contrastive Learning
EvenNet: Ignoring Odd-Hop Neighbors Improves Robustness of Graph Neural Networks
MCMAE: Masked Convolution Meets Masked Autoencoders
开场
大规模多模态预训练的最新研究进展
机器学习驱动的求解器研究
Improving 3D-aware Image Synthesis with A Geometry Dis criminator
PIE-G：Pretrained Image Encoder for Generalizable Visual Reinforcement Learning
Rethinking Lipschitz Neural Networks and Certified Robustness: A Boolean Function Perspective
Learning Neural Set Functions Under the Optimal Subset Oracle
TalentAI企业招聘
Towards Open-world Reinforcement Learning
数据高效的深度视觉识别
Jump Self-attention: Capturing High-order Statistics in Transformers
Multiagent Q-learning with Sub-Team Coordination
Convolutional Nerual Networks on Graphs with Chebyshev Approximation, Revisited
A Policy-Guided Imitation Approach for Offline Reinforcement Learning
Uncovering the Structural Fairness in Graph Contrastive Learning
EvenNet: Ignoring Odd-Hop Neighbors Improves Robustness of Graph Neural Networks
MCMAE: Masked Convolution Meets Masked Autoencoders
