activity
20182021
most citedHarnessing the Vulnerability of Latent Layers in Adversarially Trained Models

23 citations · 43 across the 8 of their papers we have counts for

collaborators

11 papers

cs.LG20211 cited

D2C: Diffusion-Denoising Models for Few-shot Conditional Generation

Abhishek Sinha, Jiaming Song, Chenlin Meng +1

Conditional generative models of high-dimensional images have many applications, but supervision signals from conditions to images can be expensive to acquire. This paper describes…

cs.CV20218 cited

Negative Data Augmentation

Abhishek Sinha, Kumar Ayush, Jiaming Song +3

Data augmentation is often used to enlarge datasets with synthetic samples generated in accordance with the underlying data distribution. To enable a wider range of augmentations,…

cs.CV20205 cited

On the Benefits of Models with Perceptually-Aligned Gradients

Gunjan Aggarwal, Abhishek Sinha, Nupur Kumari +1

Adversarial robust models have been shown to learn more robust and interpretable features than standard trained models. As shown in [\cite{tsipras2018robustness}], such robust mode…

cs.AI20201 cited

Inducing Cooperative behaviour in Sequential-Social dilemmas through Multi-Agent Reinforcement Learning using Status-Quo Loss

Pinkesh Badjatiya, Mausoom Sarkar, Abhishek Sinha +4

In social dilemma situations, individual rationality leads to sub-optimal group outcomes. Several human engagements can be modeled as a sequential (multi-step) social dilemmas. How…

cs.CV2019

cFineGAN: Unsupervised multi-conditional fine-grained image generation

Gunjan Aggarwal, Abhishek Sinha

We propose an unsupervised multi-conditional image generation pipeline: cFineGAN, that can generate an image conditioned on two input images such that the generated image preserves…

cs.LG20195 cited

A Method for Computing Class-wise Universal Adversarial Perturbations

Tejus Gupta, Abhishek Sinha, Nupur Kumari +2

We present an algorithm for computing class-specific universal adversarial perturbations for deep neural networks. Such perturbations can induce misclassification in a large fracti…