一览 CMU 的 33 篇 ICML 2019 论文

一览 CMU 的 33 篇 ICML 2019 论文

雷锋网 AI 科技评论按:机器学习顶会 ICML 2019 近期公布了接收论文清单,多家企业研究院也按惯例介绍了自己的录用论文。不过新鲜的是,今年卡耐基梅隆大学(CMU)也专门发博客列出了来自 CMU 师生的录用论文,一共有 33 篇之多。其实 UC 伯克利、斯坦福、CMU 等计算机领域实力很强的高校每年都有许多顶会论文,多样性和深度也都很好,但是宣传力度往往比不上企业级 AI 研究机构。所以 CMU 这篇博客也提醒了各位不要忘了顶级高校的学术实力,到时参会的研究人员也可以到他们的海报前一期讨论这些前沿研究进展。

雷锋网 AI 科技评论把论文列表简单介绍如下。

  • Statistical Foundations of Virtual Democracy

  • TarMAC: Targeted Multi-Agent Communication

  • A Kernel Theory of Modern Data Augmentation

  • Myopic Posterior Sampling for Adaptive Goal Oriented Design of Experiments

    • 用于适应性目标向的实验设计的短视后验抽样

  • Nearest neighbor and kernel survival analysis: Nonasymptotic error bounds and strong consistency rates

    • 最近邻以及核生存方法的分析:非渐近性的错误边界以及强相关性率

  • Policy Certificates: Towards Accountable Reinforcement Learning

  • Deep Counterfactual Regret Minimization

  • Domain Adaptation with Asymmetrically-Relaxed Distribution Alignment

  • Provably efficient RL with Rich Observations via Latent State Decoding

  • A Baseline for Any Order Gradient Estimation in Stochastic Computation Graphs

    • 可以用于任意阶的随机计算图的梯度估计的基线方法

  • Gradient Descent Finds Global Minima of Deep Neural Networks

  • Stable-Predictive Optimistic Counterfactual Regret Minimization

  • Regret Circuits: Composability of Regret Minimizers

  • Provable Guarantees for Gradient-Based MetaLearning

  • Dimensionality Reduction for Tukey Regression

    • 土耳其回归方法的降维

  • Certified Adversarial Robustness via Randomized Smoothing

  • Provably Efficient Imitation Learning from Observation Alone

    • 证明仅仅从观察就可以实现高效的模仿学习

  • SATNet: Bridging deep learning and logical reasoning using a differentiable satisfiability solver

    • SATNet:通过可微分可满足性解算器连接深度学习和逻辑推理

  • Collective Model Fusion for Multiple Black-Box Experts

  • Fine-Grained Analysis of Optimization and Generalization for Overparameterized Two-Layer Neural Networks

  • Width Provably Matters in Optimization for Deep Linear Neural Networks

  • Causal Discovery and Forecasting in Nonstationary Environments with State-Space Models

    • 在非固定环境下用状态空间模型进行因素发现和预测

  • Uniform Convergence Rate of the Kernel Density Estimator Adaptive to Intrinsic Volume Dimension

  • Faster Algorithms for Boolean Matrix Factorization

    • 更快的布尔矩阵分解算法

  • Contextual Memory Trees

  • Fault Tolerance in Iterative-Convergent Machine Learning

  • Wasserstein Adversarial Examples via Projected Sinkhorn Iterations

  • Learning to Explore via Disagreement

    • 通过不一致学习探索

  • What is the Effect of Importance Weighting in Deep Learning?

  • Adversarial camera stickers: A physical camera-based attack on deep learning systems

  • On Learning Invariant Representation for Domain Adaptation

  • Finding Options that Minimize Planning Time

  • Tight Kernel Query Complexity of Kernel Ridge

    • 核岭回归的核查询复杂度分析

ICML 2019 将于今年 6 月 10 日至 15 日在美国加州长滩举行。雷锋网(公众号:雷锋网) AI 科技评论到时也会进行多方位的报道,敬请继续关注。


一览 CMU 的 33 篇 ICML 2019 论文

原创文章,作者:ItWorker,如若转载,请注明出处:https://blog.ytso.com/135245.html

(0)
上一篇 2021年9月1日
下一篇 2021年9月1日

相关推荐

发表回复

登录后才能评论