Publications

Publication

January 05, 2021

IR-VIC: Unsupervised Discovery of Sub-goals for Transfer in RL | Facebook AI Research

We propose a novel framework to identify subgoals useful for exploration in sequential decision making tasks under partial observability. We utilize the variational intrinsic control…

Nirbhay Modhe, Prithvijit Chattopadhyay, Mohit Sharma, Abhishek Das, Devi Parikh, Dhruv Batra, Ramakrishna Vedantam

January 05, 2021

Read the Paper

Publication

January 01, 2021

Asynchronous Gradient-Push | Facebook AI Research

We consider a multi-agent framework for distributed optimization where each agent has access to a local smooth strongly convex function, and the collective goal is to achieve consensus on the parameters that minimize the sum of the agents’…

Mahmoud Assran, Michael Rabbat

January 01, 2021

Read the Paper

Publication

October 26, 2020

NLP

Weak-Attention Suppression For Transformer Based Speech Recognition

Transformers, originally proposed for natural language processing (NLP) tasks, have recently achieved great success in automatic…

Yangyang Shi, Yongqiang Wang, Chunyang Wu, Christian Fuegen, Frank Zhang, Duc Le, Ching-Feng Yeh, Michael L. Seltzer

October 26, 2020

Read the Paper

Publication

October 25, 2020

NLP

Statistical Testing on ASR Performance via Blockwise Bootstrap

A common question being raised in automatic speech recognition (ASR) evaluations is how reliable is an observed word error rate (WER) improvement comparing two ASR systems, where statistical hypothesis testing and confidence interval…

Zhe Liu, Fuchun Peng

October 25, 2020

Read the Paper

Publication

October 18, 2020

Beyond English-Centric Multilingual Machine Translation

In this work, we create a true Many-to-Many multilingual translation model that can translate directly between any pair of 100 languages.

Angela Fan, Shruti Bhosale, Holger Schwenk, Zhiyi Ma, Ahmed El-Kishky, Siddharth Goyal, Mandeep Baines, Onur Celebi, Guillaume Wenzek, Vishrav Chaudhary, Naman Goyal, Tom Birch, Vitaliy Liptchinsky, Sergey Edunov, Edouard Grave, Michael Auli, Armand Joulin

October 18, 2020

Read the Paper

Publication

October 01, 2020

Neural Relational Autoregression for High-Resolution COVID-19 Forecasting

The prevalent approach to neural machine translation relies on bi-directional LSTMs to encode the source sentence. We present a faster and simpler architecture based on a succession of convolutional layers. This allows to encode the source…

Matthew Le, Mark Ibrahim, Levent Sagun, Timothee Lacroix, Maximilian Nickel

October 01, 2020

Read the Paper

Publication

October 01, 2020

Voice Separation with an Unknown Number of Multiple Speakers

We present a new method for separating a mixed audio sequence, in which multiple voices speak simultaneously. The new method employs gated neural networks that are trained to separate the voices at multiple processing steps…

Eliya Nachmani, Yossi Adi, Lior Wolf

October 01, 2020

Read the Paper

Publication

September 14, 2020

Machine Learning in Compilers: Past, Present, and Future

In this paper we will give a retrospective of machine learning in compiler optimisation from its earliest inception, through some of the works that set themselves apart, to today’s deep learning, finishing with our vision of the field’s future.

Hugh Leather, Chris Cummins

September 14, 2020

Read the Paper

Publication

September 07, 2020

COMPUTER VISION

Predicting A Creator’s Preferences In, and From, Interactive Generative Art

As a lay user creates an art piece using an interactive generative art tool, what, if anything, do the choices they make tell us about them and their preferences? Both within the generative art form, and otherwise? As a preliminary study, we collect preferences from 311 subjects, in a specific generative art form and in other walks of life.

Devi Parikh

September 07, 2020

Read the Paper

Publication

September 07, 2020

COMPUTER VISION

Exploring Crowd Co-creation Scenarios for Sketches

As a first step towards studying the ability of human crowds and machines to effectively co-create, we explore several human-only collaborative co-creation scenarios. The goal in each scenario is to create a digital sketch using a simple web interface.

Devi Parikh, Larry Zitnick

September 07, 2020

Read the Paper

Publication

September 07, 2020

Lemotif: An Affective Visual Journal Using Deep Neural Networks

processing and image generation system that uses machine learning to (1) parse a text-based input journal entry describing the user’s day for salient themes and emotions and (2) visualize the detected themes and emotions in creative and appealing image motifs.

Devi Parikh,, X. Alice Li

September 07, 2020

Read the Paper

Publication

September 07, 2020

COMPUTER VISION

Feel The Music: Automatically Generating A Dance For An Input Song

We present a general computational approach that enables a machine to generate a dance for any input music. We encode intuitive, flexible heuristics for what a ‘good’ dance is: the structure of the dance should align with the structure of the music.