no code implementations • EMNLP 2021 • Vivek Madan, Ashish Khetan, Zohar Karnin
In this paper, we address the problem for the case when the downstream corpus is too small for additional pre-training.
no code implementations • 23 May 2022 • Anastasia Razdaibiedina, Ashish Khetan, Zohar Karnin, Daniel Khashabi, Vishaal Kapoor, Vivek Madan
In this paper, we propose Representation Projection Invariance (REPINA), a novel regularization method to maintain the information content of representation and reduce representation collapse during fine-tuning by discouraging undesirable changes in the representations.
no code implementations • ACL 2022 • Xin Huang, Ashish Khetan, Rene Bidart, Zohar Karnin
Transformer-based language models such as BERT have achieved the state-of-the-art performance on various NLP tasks, but are computationally prohibitive.
no code implementations • 1 Jan 2021 • Vivek Madan, Ashish Khetan, Zohar Karnin
The need for such a method is clear as it is infeasible to collect a large pre-training corpus for every possible domain.
12 code implementations • 11 Dec 2020 • Xin Huang, Ashish Khetan, Milan Cvitkovic, Zohar Karnin
We propose TabTransformer, a novel deep tabular data modeling architecture for supervised and semi-supervised learning.
no code implementations • 22 May 2020 • Ashish Khetan, Zohar Karnin
The methods that start with a pretrained network either prune channels uniformly across the layers or prune channels based on the basic statistics of the network parameters.
no code implementations • ACL 2020 • Ashish Khetan, Zohar Karnin
In this work we revisit the architecture choices of BERT in efforts to obtain a lighter model.
no code implementations • ICLR 2020 • Shashank Singh, Ashish Khetan, Zohar Karnin
In many learning situations, resources at inference time are significantly more constrained than resources at training time.
no code implementations • 9 Jun 2019 • Kiran Koshy Thekumparampil, Sewoong Oh, Ashish Khetan
Matching the performance of conditional Generative Adversarial Networks with little supervision is an important task, especially in venturing into new domains.
no code implementations • 20 May 2019 • Shashank Singh, Ashish Khetan, Zohar Karnin
In many learning situations, resources at inference time are significantly more constrained than resources at training time.
no code implementations • 1 Dec 2018 • Ashish Khetan, Harshay Shah, Sewoong Oh
This representation is crucial in introducing a novel estimator for the number of connected components for general graphs, under the knowledge of the spectral gap of the original graph.
2 code implementations • NeurIPS 2018 • Kiran Koshy Thekumparampil, Ashish Khetan, Zinan Lin, Sewoong Oh
When the distribution of the noise is known, we introduce a novel architecture which we call Robust Conditional GAN (RCGAN).
1 code implementation • ICLR 2018 • Ashish Khetan, Zachary C. Lipton, Anima Anandkumar
We propose a new algorithm for jointly modeling labels and worker quality from noisy crowd-sourced data.
7 code implementations • NeurIPS 2018 • Zinan Lin, Ashish Khetan, Giulia Fanti, Sewoong Oh
Generative adversarial networks (GANs) are innovative techniques for learning generative models of complex data distributions from samples.
1 code implementation • NeurIPS 2017 • Ashish Khetan, Sewoong Oh
This paper focuses on the technical challenges in accurately estimating the Schatten norms from a sampling of a matrix.
no code implementations • 18 Mar 2017 • Ashish Khetan, Sewoong Oh
We propose first estimating the Schatten $k$-norms of a matrix, and then applying Chebyshev approximation to the spectral sum function or applying moment matching in Wasserstein distance to recover the singular values.
no code implementations • NeurIPS 2016 • Ashish Khetan, Sewoong Oh
For massive and heterogeneous modern datasets, it is of fundamental interest to provide guarantees on the accuracy of estimation when computational resources are limited.
no code implementations • NeurIPS 2016 • Ashish Khetan, Sewoong Oh
Under this generalized Dawid-Skene model, we characterize the fundamental trade-off between budget and accuracy.
no code implementations • 21 Jan 2016 • Ashish Khetan, Sewoong Oh
Rank aggregation systems collect ordinal preferences from individuals to produce a global ranking that represents the social preference.