I am Research Scientist at Facebook.
My research is profiling and optimizing machine learning and deep learning applications on parallel processors such as CPU, GPU, or Xeon Phi, utilizing TensorFlow/PyTorch framework. From the analysis, I identify the bottlenecks in both architecture and software level and suggest HW/SW optimizations. Recently, I have been working on optimizing the CNN/RNN applications in the architecture level.
My previous researches were handheld system design (mobile phones/tablets and general IP architectures), and CGRA acceleration.
GPA: 3.59/4.3
Percentage Equivalent: 92
Jihyun Ryoo, Mahmut T. Kandemir, Mustafa Karakoy
Memory Space Recycling.
In Proceedings of the ACM on Measurement and Analysis of Computing Systems (POMACS) / Sigmetrics. March 2022.
[bibtex]
Mahmut T. Kandemir, Xulong Tang, Hui Zhao, Jihyun Ryoo, Mustafa Karakoy
Distance-in-Time verse Distance-in-Space.
In Proceedings of the ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI). June 2021.
[bibtex]
Mahmut T. Kandemir, Jihyun Ryoo, Xulong Tang, Mustafa Karakoy
Compiler Support for Near Data Computing.
In Proceedings of the ACM SIGPLAN Conference on Principles and Practice of Parallel Programming (PPoPP). Feb 2021.
[bibtex]
Huaipan Jiang, Anup Sarma, Mengran Fan, Jihyun Ryoo, Meena Arunachalam, Sharada Naveen, Mahmut T. Kandemir
Morphable Convolutional Neural Network for Biomedical Image Segmentation.
(Poster Present) In Proceedings of IEEE International Conference on Design, Automation and Test In Europe Conference (DATE). Feb 2021.
[bibtex]
Mahmut T. Kandemir, Jihyun Ryoo, Hui Zhao, Myoungsoo Jung, Mustafa Karakoy
Collective Affinity Aware Computation Mapping.
(Poster Present) In Proceedings of the ACM International Conference on Parallel Architectures and Compilation Techniques (PACT). Sep 2020.
[bibtex]
Jihyun Ryoo, Mengran Fan, Xulong Tang, Huaipan Jiang, Meena Arunachalam, Sharada Naveen, Mahmut T. Kandemir
Architecture-Centric Bottleneck Analysis for Deep Neural Network Applications.
In Proceedings of IEEE International Conference on High Performance Computing, Data, and Analytics (HiPC). Dec 2019.
[bibtex]
Jihyun Ryoo, Orhan Kislal, Xulong Tang, Mahmut T. Kandemir
Quantifying and Optimizing Data Access Parallelism on Manycores.
In Proceedings of IEEE International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems (MASCOTS). Sep 2018.
[bibtex]
Huaipan Jiang, Anup Sarma, Jihyun Ryoo, Jagadish B. Kortra, Meena Arunachalam, Chita R. Das, Mahmut T. Kandemir
A Learning-Guided Hierarchical Approach for Biomedical Image Segmentation.
In Proceedings of IEEE International System-on-Chip Conference (SOCC). Sep 2018.
[bibtex]
Jihyun Ryoo, Meena Arunachalam, Rahul Khanna, Mahmut T. Kandemir
Efficient K nearest neighbor algorithm implementations for throughput-oriented architectures.
In Proceedings of IEEE International Symposium on Quality Electronic Design (ISQED). March 2018.
[bibtex]
Nachiappan Chidambaram Nachiappan, Haibo Zhang, Jihyun Ryoo, Niranjan Soundararajan, Anand Sivasubramaniam, Mahmut T. Kandemir, Ravi Iyer, Chita R. Das
VIP: Virtualizing IP chains on handheld platforms.
In Proceedings of IEEE International Symposium on Computer Architecture (ISCA). June 2015.
[bibtex]
Jihyun Ryoo, Kyuseung Han, Kiyoung Choi
Leveraging parallelism in the presence of control flow on CGRAs.
In Proceedings of Asia and South Pacific Design Automation Conference (ASP-DAC). Jan 2014.
[bibtex]
Jihyun Ryoo, Seuk Son, Jaeha Kim
Design of low-power high-radix switch fabric with partially-activated input and output lines.
In Proceedings of International SoC Design Conference (ISOCC). Nov 2012.
[bibtex]
Work at Ads Training Foundation team under Ads Infra ENG Group: ML.
Worked at Machine Learning Performance (MLP) team under IAGS group.
Worked at High Performance Computing (HPC) team under SSG group.
Worked at Design Automation Laboratory in Seoul National University with Dr. Kiyoung Choi.