Dhruv Mahajan

Research Scientist

I am a research scientist on the Computer Vision team, which is part of Applied Machine Learning. I joined Facebook Research in January 2017. Prior to joining Facebook, I worked at Microsoft, MSR India and Yahoo! Research.
My research interests lie in computer vision and large scale and distributed machine learning. In computer vision, I am interested in extreme classification and ranking problems with huge scale both in terms of number of data points and labels and video understanding. I am also interested in new and efficient distributed optimization algorithms respecting system constraints like communication and machine failures.


Computer vision, large scale and distributed machine learning, and extreme classification (with large number of labels)

Latest Publications

NeurIPS - December 6, 2020

Self-Supervised Learning by Cross-Modal Audio-Video Clustering

Humam Alwassel, Dhruv Mahajan, Bruno Korbar, Lorenzo Torresani, Bernard Ghanem, Du Tran

ECCV - September 1, 2020

What leads to generalization of object proposals?

Rui Wang, Dhruv Mahajan, Vignesh Ramanathan

CVPR - June 18, 2020

DLWL: Improving Detection for Lowshot classes with Weakly Labelled data

Vignesh Ramanathan, Rui Wang, Dhruv Mahajan

CVPR - June 16, 2020

From Patches to Pictures (PaQ-2-PiQ): Mapping the Perceptual Space of Picture Quality

Zhenqiang Ying, Haoran Niu, Praful Gupta, Dhruv Mahajan, Deepti Ghadiyaram, Alan Bovik

CVPR - June 14, 2020

Don’t Judge an Object by Its Context: Learning to Overcome Contextual Bias

Krishna Kumar Singh, Dhruv Mahajan, Kristen Grauman, Yong Jae Lee, Matt Feiszli, Deepti Ghadiyaram

CVPR - May 19, 2020

ClusterFit: Improving Generalization of Visual Representations

Xueting Yan, Ishan Misra, Abhinav Gupta, Deepti Ghadiyaram, Dhruv Mahajan

CVPR - May 19, 2020

Cluster and Re-Learn: Improving Generalization of Visual Representations

Xueting Yan, Ishan Misra, Abhinav Gupta, Deepti Ghadiyaram, Dhruv Mahajan

ICCV - October 27, 2019

Scaling and Benchmarking Self-Supervised Visual Representation Learning

Priya Goyal, Dhruv Mahajan, Abhinav Gupta, Ishan Misra

CVPR - May 1, 2019

Large-scale weakly-supervised pre-training for video action recognition

Deepti Ghadiyaram, Matt Feiszli, Du Tran, Xueting Yan, Hen Wang, Dhruv Mahajan

CVPR 2018 - June 18, 2018

What Makes a Video a Video: Analyzing Temporal Information in Video Understanding Models and Datasets

De-An Huang, Vignesh Ramanathan, Dhruv Mahajan, Lorenzo Torresani, Manohar Paluri, Li Fei-Fei, Juan Carlos Niebles

ArXiv - May 2, 2018

Exploring the Limits of Weakly Supervised Pretraining

Dhruv Mahajan, Ross Girshick, Vignesh Ramanathan, Kaiming He, Manohar Paluri, Yixuan Li, Ashwin Bharambe, Laurens van der Maaten