318 citations
- Carnegie Mellon UniversityUS3 papers
- Apple (United States)US2 papers
- Amazon (Germany)DE1 paper
- Duke UniversityUS1 paper
- Google (United States)US1 paper
- Karlsruhe Institute of TechnologyDE1 paper
- Martin Luther University Halle-WittenbergDE1 paper
- Microsoft (Finland)FI1 paper
- Microsoft Research (United Kingdom)GB1 paper
- Moscow Institute of Thermal TechnologyRU1 paper
- Princeton UniversityUS1 paper
- Stanford UniversityUS1 paper
6 papers · 1 filter
Private Adaptive Gradient Methods for Convex Optimization
Hilal Asi, John Duchi, Alireza Fallah +2
We study adaptive methods for differentially private convex optimization, proposing and analyzing differentially private variants of a Stochastic Gradient Descent (SGD) algorithm w…
Uncertainty Weighted Actor-Critic for Offline Reinforcement Learning
Yue Wu, Shuangfei Zhai, Nitish Srivastava +4
Offline Reinforcement Learning promises to learn effective policies from previously-collected, static datasets without the need for exploration. However, existing Q-learning and ac…
MetricOpt: Learning to Optimize Black-Box Evaluation Metrics
Chen Huang, Shuangfei Zhai, Pengsheng Guo +1
We study the problem of directly optimizing arbitrary non-differentiable task evaluation metrics such as misclassification rate and recall. Our method, named MetricOpt, operates in…
Lossless Compression of Efficient Private Local Randomizers
Vitaly Feldman, Kunal Talwar
Locally Differentially Private (LDP) Reports are commonly used for collection of statistics and machine learning in the federated setting. In many cases the best known LDP algorith…
Dynamic curriculum learning via data parameters for noise robust keyword spotting
Takuya Higuchi, Shreyas Saxena, Mehrez Souden +3
We propose dynamic curriculum learning via data parameters for noise robust keyword spotting. Data parameter learning has recently been introduced for image processing, where weigh…
Whispered and Lombard Neural Speech Synthesis
Qiong Hu, Tobias Bleisch, Petko Petkov +3
It is desirable for a text-to-speech system to take into account the environment where synthetic speech is presented, and provide appropriate context-dependent output to the user.…