World's Best Scientists 2026 revealed!

D-Index & Metrics

Computer Science

D-Index
34
Citations
6313
World Ranking
12024
National Ranking
4906

Overview

Mikhail Smelyanskiy is affiliated with Nvidia in the United States, focusing on research within the field of Computer Science. Their work spans multiple subfields including Computer Networks and Communications, Hardware and Architecture, Artificial Intelligence, Computer Vision and Pattern Recognition, and Information Systems. The breadth of research topics covered reflects an interdisciplinary approach to advancements in computing technologies.

Their scholarly output includes a series of recent papers that address various aspects of deep learning systems, inference architectures, and scalable training methods. Notable publications include:

  • Deep Learning Training in Facebook Data Centers: Design of Scale-up and Scale-out Systems (2020, arXiv [Cornell University])
  • FBGEMM: Enabling High-Performance Low-Precision Deep Learning Inference (2021, arXiv [Cornell University])
  • Supporting Massive DLRM Inference through Software Defined Memory (2022, 2022 IEEE 42nd International Conference on Distributed Computing Systems [ICDCS])
  • Low-Precision Hardware Architectures Meet Recommendation Model Inference at Scale (2021, IEEE Micro)
  • Software-Hardware Co-design for Fast and Scalable Training of Deep Learning Recommendation Models (2021, arXiv [Cornell University])

The topics investigated in these papers highlight main research themes such as advanced data storage technologies, parallel computing and optimization techniques, advanced neural network applications, recommender systems and techniques, caching and content delivery, stochastic gradient optimization techniques, and machine learning and data classification. These themes indicate a focus on the infrastructural and algorithmic challenges in deploying and optimizing machine learning at scale.

Smelyanskiy frequently collaborates with several co-authors, contributing to a diverse research network. Regular collaborators include Maxim Naumov, Changkyu Kim, Hector Yuen, Dheevatsa Mudigere, and Jongsoo Park.

Research dissemination primarily occurs through venues such as arXiv (Cornell University), which accounts for multiple publications, the 2022 IEEE 42nd International Conference on Distributed Computing Systems (ICDCS), and IEEE Micro. These publication venues suggest a balance between preprint dissemination and peer-reviewed conferences and journals focused on computing systems.

Best Publications

  • On Large-Batch Training for Deep Learning: Generalization Gap and Sharp Minima

    Nitish Shirish Keskar;Dheevatsa Mudigere;Jorge Nocedal;Mikhail Smelyanskiy

  • Debunking the 100X GPU vs. CPU myth: an evaluation of throughput computing on CPU and GPU

    Victor W. Lee;Changkyu Kim;Jatin Chhugani;Michael Deisher

  • Applied Machine Learning at Facebook: A Datacenter Infrastructure Perspective

    Kim Hazelwood;Sarah Bird;David Brooks;Soumith Chintala

  • On Large-Batch Training for Deep Learning: Generalization Gap and Sharp Minima

    Nitish Shirish Keskar;Dheevatsa Mudigere;Jorge Nocedal;Mikhail Smelyanskiy

  • Efficient sparse matrix-vector multiplication on x86-based many-core processors

    Xing Liu;Mikhail Smelyanskiy;Edmond Chow;Pradeep Dubey

  • Debunking the 100X GPU vs. CPU myth

    Unknown

  • The Architectural Implications of Facebook's DNN-Based Personalized Recommendation

    Udit Gupta;Carole-Jean Wu;Xiaodong Wang;Maxim Naumov

  • Design and Implementation of the Linpack Benchmark for Single and Multi-node Systems Based on Intel® Xeon Phi Coprocessor

    Alexander Heinecke;Karthikeyan Vaidyanathan;Mikhail Smelyanskiy;Alexander Kobotov

  • Exploring SIMD for Molecular Dynamics, Using Intel® Xeon® Processors and Intel® Xeon Phi Coprocessors

    S. J. Pennycook;C. J. Hughes;M. Smelyanskiy;S. A. Jarvis

  • RecNMP: accelerating personalized recommendation with near-memory processing

    Liu Ke;Udit Gupta;Benjamin Youngjae Cho;David Brooks

  • Petascale high order dynamic rupture earthquake simulations on heterogeneous supercomputers

    Alexander Heinecke;Alexander Breuer;Sebastian Rettenberger;Michael Bader

  • qHiPSTER: The Quantum High Performance Software Testing Environment

    Mikhail Smelyanskiy;Nicolas P. D. Sawaya;Alán Aspuru-Guzik

  • Anatomy of High-Performance Many-Threaded Matrix Multiplication

    Tyler M. Smith;Robert van de Geijn;Mikhail Smelyanskiy;Jeff R. Hammond

  • Convergence of Recognition, Mining, and Synthesis Workloads and Its Implications

    Yen-Kuang Chen;J. Chhugani;P. Dubey;C.J. Hughes

  • Practical optimization for hybrid quantum-classical algorithms

    Gian Giacomo Guerreschi;Mikhail Smelyanskiy

  • Deep Learning Inference in Facebook Data Centers: Characterization, Performance Optimizations and Hardware Implications

    Jongsoo Park;Maxim Naumov;Protonu Basu;Summer Deng

  • Can traditional programming bridge the Ninja performance gap for parallel computing applications

    Nadathur Satish;Changkyu Kim;Jatin Chhugani;Hideki Saito

  • The BLIS Framework: Experiments in Portability

    Field G. Van Zee;Tyler M. Smith;Bryan Marker;Tze Meng Low

  • Mapping High-Fidelity Volume Rendering for Medical Imaging to CPU, GPU and Many-Core Architectures

    M. Smelyanskiy;D. Holmes;J. Chhugani;A. Larson

  • Software-Hardware Co-design for Fast and Scalable Training of Deep Learning Recommendation Models

    Dheevatsa Mudigere;Yuchen Hao;Jianyu Huang;Zhihao Jia

  • Vector instructions to enable efficient synchronization and parallel reduction operations

    Mikhail Smelyanskiy;Sanjeev Kumar;Daehyun Kim;Jatin Chhugani

  • Sparsifying Synchronization for High-Performance Shared-Memory Sparse Triangular Solver

    Jongsoo Park;Mikhail Smelyanskiy;Narayanan Sundaram;Pradeep Dubey

  • Optimization of geometric multigrid for emerging multi- and manycore processors

    Samuel Williams;Dhiraj D. Kalamkar;Amik Singh;Anand M. Deshpande

  • Deep Learning Training in Facebook Data Centers: Design of Scale-up and Scale-out Systems.

    Maxim Naumov;John Kim;Dheevatsa Mudigere;Srinivas Sridharan

  • The Architectural Implications of Facebook's DNN-based Personalized Recommendation

    Udit Gupta;Carole-Jean Wu;Xiaodong Wang;Maxim Naumov

Frequent Co-Authors

Pradeep Dubey
Pradeep Dubey Intel (United States)
Changkyu Kim
Changkyu Kim Facebook (United States)
Nadathur Satish
Nadathur Satish Facebook (United States)
Yen-Kuang Chen
Yen-Kuang Chen Alibaba Group (China)
Edward S. Davidson
Edward S. Davidson University of Michigan–Ann Arbor
Hsien-Hsin S. Lee
Hsien-Hsin S. Lee Intel (United States)
Carole-Jean Wu
Carole-Jean Wu Meta Platforms, Inc.
Kim Hazelwood
Kim Hazelwood Facebook (United States)
David Brooks
David Brooks Harvard University
John L. Volakis
John L. Volakis Florida International University

If you think any of the details on this page are incorrect, let us know.

Report an issue

We appreciate your kind effort to assist us to improve this page, it would be helpful providing us with as much detail as possible in the text box below:

Related Online Degrees & Career Pathways

Exploring related online degrees can open up many career pathways beyond traditional computer science. Students interested in a broader science foundation may consider an online bachelor’s degree in physics for its strong analytical and problem-solving components.

Data science is rapidly growing and offers excellent salaries. For prospective students looking to maximize value, researching an affordable data science degree can be a strategic choice, especially with the demand for experts in big data and machine learning.

Engineering skills are always in demand. Those considering online programs should review electrical engineering online tuition costs to find quality education that fits their budget.

If you’re seeking a quicker route into the tech industry, there are quick certifications that pay well. These can provide a fast track to high-earning roles without the time commitment of a full degree.

Best Scientists Citing Mikhail Smelyanskiy

Trending Scientists