Understanding Deep Neural Networks with Rectified Linear Units
1 version
RA
Raman Arora
cs.LGstat.MLcs.AImath.OCcs.CRcs.GTcs.CLcs.CVcs.DCcs.RO
On Valency
published · living versionsPreprints & journals
56 papers in the corpus · 2008–2026Differential Privacy of Gradient Descent on Perturbed Objectives
0 citations
Exact Unlearning in Reinforcement Learning
0 citations
Minimax-Optimal Policy Regret in Partially Observable Markov Games
0 citations
On The Statistical Complexity of Offline Decision-Making
0 citations
Learning in Markov Games with Adaptive Adversaries: Policy Regret, Fundamental Barriers, and Efficient Algorithms
0 citations
Offline Multitask Representation Learning for Reinforcement Learning
0 citations
Public-data Assisted Private Stochastic Optimization: Power and Limitations
0 citations
DART: A Principled Approach to Adversarially Robust Unsupervised Domain Adaptation
2 citations
On Sample-Efficient Offline Reinforcement Learning: Data Diversity, Posterior Sampling, and Beyond
0 citations
Differentially Private Non-Convex Optimization under the KL Condition with Optimal Rates
0 citations
From Adaptive Query Release to Machine Unlearning
0 citations
VIPeR: Provably Efficient Algorithm for Offline RL with Neural Function Approximation
2 citations
On Instance-Dependent Bounds for Offline Reinforcement Learning with Linear Function Approximation
5 citations
A Risk-Sensitive Approach to Policy Optimization
6 citations
Adversarial Robustness is at Odds with Lazy Training
3 citations
Faster Rates of Convergence to Stationary Points in Differentially Private Optimization
3 citations
Differentially Private Generalized Linear Models Revisited
4 citations
Machine Unlearning via Algorithmic Stability
26 citations
On Convergence and Generalization of Dropout Training
15 citations
Adversarial Robustness of Supervised Sparse Coding
10 citations
FetchSGD: Communication-Efficient Federated Learning with Sketching
234 citations
Is Network the Bottleneck of Distributed Training?
76 citations
Corralling Stochastic Bandit Algorithms
4 citations
Dropout: Explicit Forms and Capacity Control
13 citations
Private Stochastic Convex Optimization: Efficient Algorithms for Non-smooth Objectives
3 citations
Multiview Representation Learning for a Union of Subspaces
1 citation
Bandits with Feedback Graphs and Switching Costs
12 citations
On Dropout and Nuclear Norm Regularization
16 citations
Communication-efficient distributed SGD with Sketching
86 citations
Learning from Multiview Correlations in Open-Domain Videos
13 citations
Policy Regret in Repeated Games
4 citations
Enter the Matrix: Factorization Uncovers Knowledge from Omics.
234 citations
Streaming Kernel PCA with Random Features
1 citation
On the Implicit Bias of Dropout
39 citations
Visual Robot Task Planning
57 citations
Learning to Imagine Manipulation Goals for Robot Task Planning
1 citation
Identifying Nominals with No Head Match Co-references Using Deep Learning
2 citations
Stochastic Approximation for Canonical Correlation Analysis
20 citations
Deep Generalized Canonical Correlation Analysis
145 citations
Symmetry, Saddle Points, and Global Optimization Landscape of Nonconvex Matrix Factorization
44 citations
The Physical Systems Behind Optimization Algorithms
10 citations
Understanding Deep Neural Networks with Rectified Linear Units
372 citations
On Faster Convergence of Cyclic Block Coordinate Descent-type Methods for Strongly Convex Minimization
6 citations
Disease Trajectory Maps
16 citations
On Fast Convergence of Proximal Algorithms for SQRT-Lasso Optimization: Don't Worry About Its Nonsmooth Loss Function
4 citations
Nonconvex Sparse Learning via Stochastic Optimization with Progressive Variance Reduction
30 citations
Embedding Lexical Features via Low-Rank Tensors
8 citations
On Deep Multi-View Representation Learning: Objectives and Optimization
51 citations
Stochastic Optimization for Deep CCA via Nonlinear Orthogonal Iterations
59 citations
Accelerated Mini-batch Randomized Block Coordinate Descent Method.
50 citations
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in beta. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.