paper.dinq.meAI 与前沿科学顶会 · 学术新星雷达

更新于

返回新星榜

MM

Michael Muehlebach

ICML 2026×6效率与系统推理强化学习与后训练理论与学习动力学检索与知识

新星分资深

35

一作贡献0

核心/参与33

方向广度10

合作网络6

0 一作 · 6 篇 · ICML 2026

获取完整研究档案

首次获取该候选人的完整档案及 StartLux 评估共 20 积分；每位用户单独解锁，之后查看免费。主动刷新 20 积分。

联系方式

实时查找电子邮箱，费用不超过20积分；未找到邮箱不扣费。

2026 顶会论文

6

作者位次 #5/6ICML 2026Poster· Optimization / Zero-order and Black-box Optimization

Zeroth-Order Optimization at the Edge of Stability

Minhak Song · Liang Zhang · Bingcong Li · Niao He · Michael Muehlebach · Sewoong Oh

作者位次 #2/3ICML 2026Spotlight· Deep Learning / Generative Models and Autoencoders

Efficient Diffusion Models under Nonconvex Equality and Inequality constraints via Landing

Kijung Jeon · Michael Muehlebach · Molei Tao

效率与系统OpenReview

作者位次 #7/7ICML 2026Poster· Deep Learning / Foundation Models

SALAAD: Sparse And Low-Rank Adaptation via ADMM for Large Language Model Inference

Hao Ma · Melis Ilayda Bal · Liang Zhang · Bingcong Li · Niao He · Melanie Zeilinger · Michael Muehlebach

效率与系统OpenReview

作者位次 #2/4ICML 2026Poster· Optimization / Everything Else

From Lyapunov Analysis to Algorithm Design in two-sided PL Minimax Optimization

Mansi Rankawat · Michael Muehlebach · Simon Lacoste-Julien · Damien Scieur

作者位次 #3/3ICML 2026Poster· Theory / Reinforcement Learning and Planning

Commit to the Bit: Reactive Reinforcement Learning Done Right

Onno Eberhard · Claire Vernade · Michael Muehlebach

推理强化学习与后训练理论与学习动力学OpenReview

作者位次 #5/5ICML 2026Spotlight· Theory / Reinforcement Learning and Planning

Why Linear Recurrent Memory Works in Partially Observable Reinforcement Learning

Yike Zhao · Onno Eberhard · Malek khammassi · Ali Sayed · Michael Muehlebach

推理强化学习与后训练检索与知识OpenReview

常见合作者

LZLiang Zhang×2 BLBingcong Li×2 NHNiao He×2 OEOnno Eberhard×2 MSMinhak Song SOSewoong Oh KJKijung Jeon MTMolei Tao HMHao Ma MBMelis Ilayda Bal MZMelanie Zeilinger MRMansi Rankawat