Publications

Publication

May 29, 2021

NLP

Autoregressive Entity Retrieval

Entities are at the center of how we represent and aggregate knowledge. For instance, Encyclopedias such as Wikipedia are structured by entities …

Nicola De Cao, , Gautier Izacard, Sebastian Riedel, Fabio Petroni

May 29, 2021

Read the Paper

Publication

May 21, 2021

NLP

Unsupervised Speech Recognition

We address the problem of grounding free-form textual phrases by using weak supervision from image-caption pairs. We propose a novel end-to-end model that uses caption-to-image retrieval as a “downstream” task to guide the process of phrase…

Alexei Baevski, Wei-Ning Hsu, Alexis Conneau, Michael Auli

May 21, 2021

Read the Paper

Publication

May 19, 2021

SYSTEMS RESEARCH

TT-REC: Tensor Train Compression for Deep Learning Recommendation Model Embeddings

The memory capacity of embedding tables in deep learning recommendation models (DLRMs) is increasing dramatically from tens of GBs to TBs across the industry …

Chunxing Yin, Bilge Acun, Xing Liu, Carole-Jean Wu

May 19, 2021

Read the Paper

Publication

May 15, 2021

COMPUTER VISION

RoI Tanh-polar Transformer Network for Face Parsing in the Wild

Face parsing aims to predict pixel-wise labels for facial components of a target face in an image. Existing approaches usually crop the target face from the input image …

Yiming Lin, Jie Shen, Yujiang Wang, Maja Pantic

May 15, 2021

Read the Paper

Publication

May 14, 2021

Not All Memories are Created Equal: Learning to Forget by Expiring

We propose Expire-Span, a method that learns to retain the most important information and expire the irrelevant information.…

Sainbayar Sukhbaatar, Da Ju, Spencer Poff, Stephen Roller, Arthur Szlam, Jason Weston, Angela Fan

May 14, 2021

Read the Paper

Publication

May 07, 2021

SYSTEMS RESEARCH

Ditto: Fair and robust federated learning through personalization

Fairness and robustness are two important concerns for federated learning systems. …

Tian Li, Shengyuan Hu, Ahmad Beirami, Virginia Smith

May 07, 2021

Read the Paper

Publication

May 07, 2021

CORE MACHINE LEARNING

THEORY

Tilted Empirical Risk Minimization

Empirical risk minimization (ERM) is typically designed to perform well on the average loss, which can result in estimators that are sensitive to outliers …

Tian Li, Ahmad Beirami, Maziar Sanjabi, Virginia Smith

May 07, 2021

Read the Paper

Publication

May 07, 2021

COMPUTER VISION

THEORY

Convolution Can Incur Foveation Effects

This exhibit demonstrates how boundary treatment in convolutional networks can incur foveation effects …

Jun Yuan, Bilal Alsallakh, Narine Kokhlikyan, Vivek Miglani, Orion Rebliz-Richardson

May 07, 2021

Read the Paper

Publication

May 04, 2021

HUMAN & MACHINE INTELLIGENCE

REINFORCEMENT LEARNING

Learning Invariant Representations for Reinforcement Learning without Reconstruction

We study how representation learning can accelerate reinforcement learning from rich observations, such as images, without relying either on domain knowledge or pixel-reconstruction. …