Date & Time:
May 2, 2019 4:00 pm – 5:00 pm
Location:
Crerar 390, 5730 S. Ellis Ave., Chicago, IL,
05/02/2019 04:00 PM 05/02/2019 05:00 PM America/Chicago Anshumali Shrivastava (Rice) – Hashing Algorithms for Extreme Scale Machine Learning Crerar 390, 5730 S. Ellis Ave., Chicago, IL,

Hashing Algorithms for Extreme Scale Machine Learning

In this talk, I will discuss some of my recent and surprising findings on the use of hashing algorithms for large-scale estimations. Locality Sensitive Hashing (LSH) is a hugely popular algorithm for sub-linear near neighbor search. However, it turns out that fundamentally LSH is a constant time (amortized) adaptive sampler from which efficient near-neighbor search is one of the many possibilities. Our observation adds another feather in the cap for LSH. LSH offers a unique capability to do smart sampling and statistical estimations at the cost of few hash lookups. Our observation bridges data structures (probabilistic hash tables) with efficient unbiased statistical estimations. I will demonstrate how this dynamic and efficient sampling beak the computational barriers in adaptive estimations where it is possible that we pay roughly the cost of uniform sampling but get the benefits of adaptive sampling. We will demonstrate the power of one simple idea for three favorite problems 1) Partition function estimation for large NLP models such as word2vec, 2) Adaptive Gradient Estimations for efficient SGD, 3) Sub-Linear Deep Learning with Huge Parameter Space. 

I will show the power of these randomized algorithm by introducing SLIDE system. SLIDE is an auspicious illustration of the power of smart randomized algorithms over CPUs in outperforming the best available GPU with an optimized implementation for training large neural networks.   Our evaluations on large industry-scale datasets, with some large fully connected architectures, show that training with SLIDE on a 44 core CPU is more than 2.7 times (2 hours vs. 5.5 hours) faster than the same network trained  using Tensorflow on Tesla V100 at any given accuracy level.

In the end, if time permits, we will switch to memory cost and see the power of a simple hashing that can shrink memory requirements associated with classification problems exponentially! Using our algorithms, we can train 100,000 classes with 400,000 features, on a single Titan X while only needing 5% or less memory required to store all the weights. Running a simple logistic regression on this data, the model size of 160 GB is unavoidable.

Host: Andrew Chien

Anshumali Shrivastava

Assistant Professor, Rice University

Anshumali Shrivastava is an assistant professor in the computer science department at Rice University.  His broad research interests include randomized algorithms for large-scale machine learning. In 2018, Science news named him one of the Top-10 scientists under 40 to watch.  He is a recipient of National Science Foundation CAREER Award, a Young Investigator Award from Air Force Office of Scientific Research, and machine learning research award from Amazon. His research on hashing inner products has won Best Paper Award at NIPS 2014 while his work on representing graphs got the Best Paper Award at IEEE/ACM ASONAM 2014.  Anshumali finished his Ph.D. in 2015 from Cornell University.

Related News & Events

Video

“Machine Learning Foundations Accelerate Innovation and Promote Trustworthiness” by Rebecca Willett

Jan 26, 2024
Video

Nightshade: Data Poisoning to Fight Generative AI with Ben Zhao

Jan 23, 2024
No Name

Five UChicago CS students named to Siebel Scholars Class of 2024

Oct 02, 2023
No Name

In The News: U.N. Officials Urge Regulation of Artificial Intelligence

"Security Council members said they feared that a new technology might prove a major threat to world peace."
Jul 27, 2023
No Name

UChicago Computer Scientists Bring in Generative Neural Networks to Stop Real-Time Video From Lagging

Jun 29, 2023
No Name

UChicago Team Wins The NIH Long COVID Computational Challenge

Jun 28, 2023
No Name

UChicago Assistant Professor Raul Castro Fernandez Receives 2023 ACM SIGMOD Test-of-Time Award

Jun 27, 2023
No Name

Computer Science Displays Catch Attention at MSI’s Annual Robot Block Party

Apr 07, 2023
No Name

UChicago, Stanford Researchers Explore How Robots and Computers Can Help Strangers Have Meaningful In-Person Conversations

Mar 29, 2023
Students posing at competition
No Name

UChicago Undergrad Team Places Second Overall In Regionals For World’s Largest Programming Competition

Mar 17, 2023
No Name

Postdoc Alum John Paparrizos Named ICDE Rising Star

Mar 15, 2023
No Name

New EAGER Grant to Asst. Prof. Eric Jonas Will Explore ML for Quantum Spectrometry

Mar 03, 2023
arrow-down-largearrow-left-largearrow-right-large-greyarrow-right-large-yellowarrow-right-largearrow-right-smallbutton-arrowclosedocumentfacebookfacet-arrow-down-whitefacet-arrow-downPage 1CheckedCheckedicon-apple-t5backgroundLayer 1icon-google-t5icon-office365-t5icon-outlook-t5backgroundLayer 1icon-outlookcom-t5backgroundLayer 1icon-yahoo-t5backgroundLayer 1internal-yellowinternalintranetlinkedinlinkoutpauseplaypresentationsearch-bluesearchshareslider-arrow-nextslider-arrow-prevtwittervideoyoutube