Om Thakkar
13 papers · 2018–2024 · 6 venues · across top CS/AI venues
Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.
Achievements
Jump to papers ↓+8 more ↓ Show less ↑
πΊοΈ Taxonomy Completionist (23) π Conference Polyglot Β· Silver π Academic Marathon Β· Sprint π Interdisciplinary Bridge Β· Bronze π£ Hot Topic Early Bird
π
Keyword Champion
π§
Keyword Pioneer
π
Grand Slam
π
Century Club Β· Bronze
β
The Questioner
π₯
Unstoppable Β· Bronze
ποΈ
Keyword Collector
(57)
π
Trend Setter
Top co-authors
Research topics
Machine Learning
(12)
Application Areas
(9)
Optimization & Theory Β· Machine Learning
(8)
Artificial Intelligence
(7)
Deep Learning
(4)
Learning Paradigms Β· Artificial Intelligence
(4)
Learning Types Β· Machine Learning
(4)
Core AI
(3)
Core Methods
(3)
Learning Paradigms Β· Machine Learning
(3)
Security & Privacy
(2)
Optimization & Theory Β· Deep Learning
(2)
Privacy
(2)
Mathematics & Optimization
(1)
Learning Types Β· Deep Learning
(1)
Keywords
differential privacy
(8)
federated learning
(4)
stochastic gradient descent
(2)
private model training
(2)
gradient clipping
(2)
hyperparameter optimization
(1)
automatic speech recognition
(1)
transfer learning
(1)
pac learning
(1)
linear regression
(1)
sample complexity
(1)
model forgetting
(1)
loss landscape
(1)
privacy-preserving learning
(1)
statistical query
(1)
speech recognition
(1)
vc dimension
(1)
distribution shift
(1)
empirical risk minimization
(1)
self-supervised learning
(1)
Papers
Efficiently Train ASR Models that Memorize Less and Perform Better with Per-core Clipping
Interspeech 2024
Quantifying Unintended Memorization in BEST-RQ ASR Encoders
Interspeech 2024
Measuring Forgetting of Memorized Training Examples
ICLR 2023
Why Is Public Pretraining Necessary for Private Model Training?
ICML 2023
Public Data-Assisted Mirror Descent for Private Model Training
ICML 2022
The Role of Adaptive Optimizers for Honest Private Hyperparameter Selection
AAAI 2022
Differentially Private Learning with Adaptive Clipping
NeurIPS 2021
Evading the Curse of Dimensionality in Unconstrained Private GLMs
AISTATS 2021
Practical and Private (Deep) Learning Without Sampling or Shuffling
ICML 2021
Revealing and Protecting Labels in Distributed Training
NeurIPS 2021
Guaranteed Validity for Empirical Approaches to Adaptive Data Analysis
AISTATS 2020
Privacy Amplification via Random Check-Ins
NeurIPS 2020
Model-Agnostic Private Learning
NeurIPS 2018