Posts by Collection

portfolio

Multimodal Understanding of Long-Form Video

Models that combine video, audio and text to retrieve moments, answer questions, localize actions and curate highlights in long videos.

Other Dolby Projects

HDR for Dolby Vision 2, and real-time active speaker detection for mobile devices and smart glasses.

Inverse Imaging with Deep Untrained Neural Network Priors

Untrained deep networks as image priors for compressed sensing, phase retrieval and HDR imaging, with convergence guarantees.

Adversarially Robust Learning via Entropic Regularization

ATENT: an entropy-regularized loss for training neural networks that are robust to adversarial perturbations.

Provable algorithms for machine learning

Gradient descent and alternating minimization for shallow ResNets, with linear convergence guarantees.

Compressive phase retrieval under image priors

CoPRAM and Block CoPRAM: fast, sample-efficient phase retrieval using sparsity and structured sparsity.

Fourier ptychography for super-resolution under image priors

Sub-diffraction imaging via Fourier ptychography using sparsity and low-rank priors.

Older Projects

Course and side projects from graduate school: CGI image forensics, topic modeling with NMF and sparse PCA, and compressive video segmentation.

publications

Fast, sample efficient algorithms for structured phase retrieval

Published in Advances in Neural Information Processing Systems (NIPS), 2017

Compressive phase retrieval using signal priors.

Low-rank Fourier ptychography

Published in IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2018

Using low-rank priors for efficient ptychography based video super-resolution.

Sub-diffraction imaging using Fourier ptychography and structured sparsity

Published in IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2018

Using sparsity priors for efficient ptychography based image super-resolution.

Towards Sample-Optimal Methods for Solving Random Quadratic Equations with Structure

Published in IEEE International Symposium on Information Theory (ISIT), 2018

Using signal priors for efficient compressive phase retrieval.

Sample efficient algorithms for recovering structured signals from magnitude-only measurements

Published in IEEE Transactions on Information Theory, 2019

Compressive phase retrieval using signal priors under noisy Gaussian measurements.

Linearly Convergent Algorithms for Learning Shallow Residual Networks

Published in IEEE International Symposium on Information Theory (ISIT), 2019

Linear convergence analysis for learning shallow networks with alternating minimization and gradient descent.

teaching

Teaching experience 1

Undergraduate course, University 1, Department, 2014

This is a description of a teaching experience. You can use markdown like any other post.

Teaching experience 2

Workshop, University 1, Department, 2015