Research Area
Year Published

999 Results

April 27, 2020

The Early Phase of Neural Network Training

International Conference on Learning Representations (ICLR)

Recent studies have shown that many important aspects of neural network learning take place within the very earliest iterations or epochs of training. For example, sparse, trainable sub-networks emerge (Frankle et al., 2019), gradient descent moves into a small subspace (Gur-Ari et al., 2018), and the network undergoes a critical period (Achille et al., 2019). Here we examine the changes that deep neural networks undergo during this early phase of training.

By: Jonathan Frankle, David J. Schwab, Ari Morcos

April 26, 2020

SlowMo: Improving Communication-Efficient Distributed SGD with Slow Momentum

International Conference on Learning Representations (ICLR)

Inspired by the BMUF method of Chen & Huo (2016), we propose a slow momentum (SLOWMO) framework, where workers periodically synchronize and perform a momentum update, after multiple iterations of a base optimization algorithm.

By: Jianyu Wang, Vinayak Tantia, Nicolas Ballas, Michael Rabbat

April 25, 2020

And the bit goes down: Revisiting the quantization of neural networks

International Conference on Learning Representations (ICLR)

In this paper, we address the problem of reducing the memory footprint of convolutional network architectures. We introduce a vector quantization method that aims at preserving the quality of the reconstruction of the network outputs rather than its weights.

By: Pierre Stock, Armand Joulin, Rémi Gribonval, Benjamin Graham, Hervé Jégou

April 25, 2020

Permutation Equivariant Models for Compositional Generalization in Language

International Conference on Learning Representations (ICLR)

Humans understand novel sentences by composing meanings and roles of core language components. In contrast, neural network models for natural language modeling fail when such compositional generalization is required. The main contribution of this paper is to hypothesize that language compositionality is a form of group-equivariance. Based on this hypothesis, we propose a set of tools for constructing equivariant sequence-to-sequence models.

By: Jonathan Gordon, David Lopez-Paz, Marco Baroni, Diane Bouchacourt

April 24, 2020

Chasm: A Screw Based Expressive Compact Haptic Actuator

ACM Conference on Human Factors in Computing Systems (CHI)

We present a compact broadband linear actuator, Chasm, that renders expressive haptic feedback on wearable and handheld devices. Unlike typical motor-based haptic devices with integrated gearheads, Chasm utilizes a miniature leadscrew coupled to a motor shaft, thereby directly translating the high-speed rotation of the motor to the linear motion of a nut carriage without an additional transmission.

By: Pornthep Preechayasomboon, Ali Israr, Majed Samad

April 20, 2020

Trajectory Optimization of Solar-Powered High-Altitude Long Endurance Aircraft

International Conference on Control, Automation, and Robotics (ICCAR)

Solar-powered high-altitude long endurance aircraft that harvest and store solar energy can fly indefinitely if they are able to close a 24-hour energy cycle. Perpetual endurance is possible when energy consumption does not exceed energy storage.

By: Jack Marriott, Birce Tezel, Zhang Liu, Nicolas Stier

April 20, 2020

Direction of Arrival Estimation in Highly Reverberant Environments Using Soft Time-Frequency Mask

IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA)

A recent approach to improving the robustness of sound localization in reverberant environments is based on pre-selection of time-frequency pixels that are dominated by direct sound. This approach is equivalent to applying a binary time-frequency mask prior to the localization stage. Although the binary mask approach was shown to be effective, it may not exploit the information available in the captured signal to its full extent. In an attempt to overcome this limitation, it is hereby proposed to employ a soft mask instead of the binary mask.

By: Vladimir Tourbabin, Jacob Donley, Boaz Rafaely, Ravish Mehra

April 9, 2020

Environment-aware reconfigurable noise suppression

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

The paper proposes an efficient, robust, and reconfigurable technique to suppress various types of noises for any sampling rate. The theoretical analyses, subjective and objective test results show that the proposed noise suppression (NS) solution significantly enhances the speech transmission index (STI), speech intelligibility (SI), signal-to-noise ratio (SNR), and subjective listening experience.

By: Jun Yang, Joshua Bingham

March 28, 2020

Communicating Socio-Emotional Sentiment Through Haptic Messages

IEEE Haptics Symposium

We explore how a custom-built, low degree-of-freedom, wearable haptic display may mediate the encoding and decoding of a set of complex socio-emotional messages that is sent and received by strangers, intimate partners, and even the same individual a week later.

By: Xi Laura Cang, Ali Israr

March 21, 2020

Evaluating Virtual Reality Experiences Through Participant Choices

IEEE Conference on Virtual Reality and 3D User Interfaces

When building virtual reality applications teams must choose between different configurations of the hardware and/or software aspects, and other factors, of the experience. In this paper we extend a framework for assessing how these factors contribute to quality of experience in an example evaluation. We consider how four factors related to avatar expressiveness affect quality of experience: Eye Gaze, Eye Blinking, Mouth Animation, and Microexpressions.

By: María Murcia López, Tara Collingwoode-Williams, William Steptoe, Raz Schwartz, Timothy Loving, Mel Slater