top of page
uofr_edited.jpg
university-rochester_216_edited_edited.jpg
Image (1).jpg

Tony Geng

Assistant Professor

Department of ECE

Department of Computer Science

University of Rochester

tong.geng(at)rochester.edu

  • g
  • universityofrochesterlogo_1487157387000_17343118_ver1.0

HELLO, I'M

Tony Geng

universityofrochesterlogo_1487157387000_17343118_ver1.0.webp

Assistant Professor @ University of Rochester

About

About Tony

MY BACKGROUND

Dr. Tony Geng is a tenure-track assistant professor in the ECE and CS departments of the University of Rochester (UR) and the director of UR's IntelliArch (Intelligent Architecture) Lab. He also holds secondary appointments with the Goergen Institute for Data Science. Before joining Rochester, Tony worked in the Physical & Computational Sciences Directorate (PCSD) at Pacific Northwest National Laboratory (PNNL) operated by the Department of Energy of the US government. His research interests are at the intersection of Computer Architecture & Systems, Generative AI, Graph Intelligence, and High-Performance Computing. Tony's papers have appeared in many prestigious conferences and journals e.g. ISCA, MICRO, HPCA, OSDI, ICLR, ICML, NIPS, CVPR, ICCV, AAAI, DAC, SC, TPDS, TC, and TIP.

Job Openings:

I am looking for two Ph.D. students, one PostDoc, and three interns to work on Sustainable Artificial General Intelligence (AGI), focusing on Generative AI, Graph Intelligence, and their applications. 

RESEARCH INTERESTS

Computer Architecture: GPU, Accelerators for AI, Nature-Powered Computer, FPGA

Machine Learning: Generative AI,  Graph Learning, Nature-Powered AI

Applications: Fintech, Social Media, Recommendation System, Scientific Computing, Power Grid

Selected Publications

​   2024:

  • [MICRO 2024] P.Haghi, C.Wu, Z.Azad, Y.Li, A.Gui, Y.Hao, A.Li, T.Geng: "Bridging the Gap Between LLMs and LNS with Dynamic Data Format and Architecture Codesign", 57th IEEE/ACM International Symposium on Microarchitecture.

  • [ISCA 2024] R.Song, C.Wu, C.Liu, A.Li, M.Huang, T.Geng: "DS-GL: Advancing Graph Learning via Harnessing Nature's Power within Scalable Dynamical Systems", the 51th IEEE/ACM International Symposium on Computer Architecture.

  • [ICLR 2024] C.Wu, R.Song, C.Liu, Y.Yang, A.Li, M.Huang, T.Geng: "Extending Power of Nature from Binary to Real-valued Graph Learning in Real World", The Twelfth International Conference on Learning Representations.

  • [NeurIPS 2024] R.Zeng, C.Han, Q.Wang, C.Wu, T.Geng, L.Huang, Y.Wu, D.Liu: "Visual Fourier Prompt Tuning", Thirty-eighth Conference on Neural Information Processing Systems.

  • [ICML 2024] C.Han, Y.Lu, G.Sun, J.Liang, Z.Cao, Q.Wang, Q.Guan, S.Dianat, R.Rao, T.Geng, Z.Tao, D.Liu "Prototypical Transformer As Unified Motion Learners", Forty-second International Conference on Machine Learning.

  • [TC 2024] C.Wu, A.Guo, P.Haghi, A.Li, T.Geng, M.Herbordt: "FPGA-Accelerated Range-Limited Molecular Dynamics", IEEE Transactions on Computers.

  • [Briefings in Bioinformatics 2024] Y.Wang, J.Zhao, H.Xu, C.Han, Z.Tao, D.Zhou, T.Geng, D.Liu, Z.Ji: "A systematic evaluation of computational methods for cell segmentation", Briefings in Bioinformatics (Impact Factor: 13.99).

  • [AIS 2024] Y.Li, J.Liu, X.Zhao, W.Liu, T.Geng, A.Li, X.Zhang: "Accurate and Data-Efficient Micro-XRD Phase Identification Using Multi-Task Learning: Application to Hydrothermal Fluids", Advanced Intelligent Systems Journal (Impact Factor: 7.4).

  • [SC 2024] H.Feng, B.Zhang, F.Ye, M.Si, C.Chu, J.Tian, C.Yin, Z.Deng, Y.Hao, P.Balaji, T.Geng, D.Tao: "Accelerating Communication in Deep Learning Recommendation Model Training with Dual-Level Adaptive Lossy Compression", Proceedings of the International Conference for High-Performance Computing, Networking, Storage, and Analysis.

  • [ICS 2024] P.Haghi, C.Tan, A.Guo, C.Wu, D.Liu, A.Li, A.Skjellum, T.Geng, M.Herbordt: "SmartFuse: Reconfigurable Smart Switches to Accelerate Fused Collectives in HPC Applications", the 38th ACM International Conference on Supercomputing.

    2023:

  • [DAC 2023] Z.Liu, Y.Yang, Z.Pan, A.Sharma, A.Hasan, C.Ding, A.Li, M.Huang, T.Geng: "Ising-CF: A Pathbreaking Collaborative Filtering Method Through Efficient Ising Machine Learning", The 59th Design Automation Conference.

  • [DAC 2023] Y.Luo*, C.Tan*, N.Agostini, A.Li, A.Tumeo, N.Dave, T.Geng: "ML-CGRA: An Integrated Compilation Framework to Enable Efficient Machine Learning Acceleration on CGRAs", The 59th Design Automation Conference.

  • [AAAI 2023] Z.Pan, A.Sharma, J.Hu, Z.liu, A.Li, H.Liu, M.Huang, T.Geng: "Ising-Traffic: An Ising-based Framework for Traffic Congestion Prediction with Uncertainty", Thirty-Seventh AAAI Conference on Artificial Intelligence.

  • [ICS 2023] A.Guo, Y.Hao, C.Wu, P.Haghi, Z.Pan, M.Si, D.Tao, A.Li, M.Herbordt, T.Geng: "Software-Hardware Co-design of Heterogeneous SmartNIC System for Recommendation Models Inference and Training", the 36th ACM International Conference on Supercomputing.

  • [MICRO 2023] U.Vengalam, Y.Liu, T.Geng, H.Wu, M.Huang: "Supporting Energy-Based Learning With an Ising Machine Substrate: A Case Study on RBM", the 56th IEEE/ACM International Symposium on Microarchitecture.

  • [SC 2023] C.Wu, T.Geng, A.Guo, S.Bandara, P.Haghi, C.Liu, A.Li, M.Herbordt: "FASDA: An FPGA-Aided, Scalable and Distributed Accelerator for Range-Limited Molecular Dynamics", Proceedings of the International Conference for High-Performance Computing, Networking, Storage, and Analysis.

  • [OSDI 2023] Y.Wang, B.Feng, Z.Wang, T.Geng, A.Li, K.Barker, Y.Ding: "MGG: Accelerating Graph Neural Networks with Fine-grained intra-kernel Communication-Computation Pipelining on Multi-GPU Platforms", USENIX Symposium on Operating Systems Design and Implementation.

  • [ICS 2023] P.Haghi, W.Krska, C.Tan, T.Geng, ..., A.Li, A.Skjellum, M.Herbordt: "FLASH: FPGA-Accelerated Smart Switches with GCN Case Study", the 36th ACM International Conference on Supercomputing.

  • [DAC 2023] H.Peng, ..., C.Wang, T.Geng, W.Wen, X.Xu, C.Ding: "PASNet: Polynomial Architecture Search Framework for Two-party Computation-based Secure Neural Network Deployment", The 59th Design Automation Conference.

  • [NeurIPS 2023] J.Liang, Y.Cui, Q.Wang, T.Geng, W.Wang, D.Liu: "ClusterFomer: Clustering As A Universal Visual Learner", Thirty-seventh Conference on Neural Information Processing Systems.

  • [NeurIPS 2023] H.Peng, R.Ran, ..., T.Geng, X.Xu, W.Wen, C.Ding: "LinGCN: Structural Linearized Graph Convolutional Network for Homomorphically Encrypted Inference", Thirty-seventh Conference on Neural Information Processing Systems.

  • [ICCV 2023] H.Peng, S.Huang, ..., T.Geng, K.Mahmood, W.Wen, X.Xu, C.Ding: "AutoReP: Automatic ReLU Replacement for Fast Private Network Inference", 2023 International Conference on Computer Vision.

  • [TIP 2023] D.Liu, J.Liang, T.Geng, A.Loui, T.Zhou: "Tripartite Feature Enhanced Pyramid Network for Dense Prediction", IEEE Transactions on Image Processing (Impact Factor: 10.86).

  • [CVPR 2023] Y.Lu, Q.Wang, S.Ma, T.Geng, Y.Chen, H.Chen, D.Liu: "TransFlow: Transformer as Flow Learner", Conference on Computer Vision and Pattern Recognition 2023.

​   2022:

  • [TPDS 2022] W.Sun, A.Li, T.Geng, S.Stuijk, H.Corporaal: "Dissecting Tensor Cores via Microbenchmarks: Latency, Throughput and Numerical Behaviors", IEEE Transactions on Parallel and Distributed Systems.

  • [HPCA 2022] H.You*, T.Geng*, Y.Zhang, A.Li, Y.Lin: "GCoD: Graph Convolutional Network Acceleration via Dedicated Algorithm and Accelerator Co-Design", The 28th IEEE International Symposium on HighPerformance Computer Architecture.

  • [HPCA 2022] C.Tan, N.B.Agostini, T.Geng, C.Xie, J.Li, A.Li, K.Barker, A.Tumeo: "DRIPS: Dynamic Rebalancing of Pipelined Streaming Applications on CGRAs", The 28th IEEE International Symposium on High-Performance Computer Architecture.

  • [DAC 2022] H. Peng, ..., T.Geng, ..., C.Ding: "A Length Adaptive Algorithm-Hardware Co-design of Transformer on FPGA Through Sparse Attention and Dynamic Pipelining", The 58th Design Automation Conference.

  • [ICS 2022] C.Zhang, S.Jin, T.Geng, J.Tian, A.Li, D.Tao: "Accelerating Parallel I/O Via Hardware-Algorithm Co-Designed Adaptive Lossy Compression", the 36th ACM International Conference on Supercomputing.

  • [ICS 2022] C.Tan, T.Tembe, J.Zhang, B.Fang, T.Geng, G.Wei, D.Brooks, A.Tumeo, G.Gopalakrishnan A.Li: "ASAP - Automatic Synthesis of Area-Efficient and  Precision-Aware CGRA", the 36th ACM International Conference on Supercomputing.

   2021:

  • [MICRO 2021] T.Geng, C.Wu, ..., M.Herbordt, Y.Lin, A.Li: "I-GCN: A Graph Convolutional Network Accelerator with Runtime Locality Enhancement through Islandization", the 54th IEEE/ACM International Symposium on Microarchitecture.

  • [TPDS 2021] T.Geng, T.Wang, C.Wu, Y.Li, ..., A.Li, M.Herbordt: "O3BNN-R: An Out-Of-Order Architecture for HighPerformance and Regularized BNN inference", IEEE Transactions on Parallel and Distributed Systems.

  • [TPDS 2021] C.Tan, C.Xie, T.Geng, ..., K.Barker, A.Li: "ARENA: Asynchronous Reconfigurable Accelerator Ring to Enable Data-Centric Parallel Computing", IEEE Transactions on Parallel and Distributed Systems.

  • [SC 2021] B.Feng, Y.Wang, T.Geng, A.Li, Y.Ding: "APNN-TC: Accelerating Arbitrary Precision Neural Networks on Ampere GPU Tensor Cores", Proceedings of the International Conference for High-Performance Computing, Networking, Storage, and Analysis.

  • [ICCAD 2021] Y.Zhang, H.You, Y.Fu, T.Geng, A.Li, Y.Lin: "G-CoS: GNN-Accelerator Co-Search Towards Both Better Accuracy and Efficiency", 2021 International Conference On Computer Aided Design.

  • [ICCD 2021] C.Tan, T.Geng, C.Xie, N.Agostini, J.Li, A.Li, K.Barker, A.Tumeo: "DynPaC: Coarse-Grained, Dynamic, and Partially Reconfigurable Array for Streaming Applications", the 39th IEEE International Conference on Computer Design. (Best Paper Award)

   2020:

  • [MICRO 2020] T.Geng, A.Li, T.Wang, C.Wu, Y.Li, ..., M.Herbordt: "AWB-GCN: A Hardware Accelerator of GraphConvolution-Network through Runtime Workload Rebalancing", the 53rd IEEE/ACM International Symposium on Microarchitecture.

  • [TC 2020] T.Geng*, T.Wang*, A.Li, X.Jin, M.Herbordt: "FPDeep: Scalable Acceleration of CNN Training on DeeplyPipelined FPGA Clusters", IEEE Transactions on Computers. 

  • [ICS 2020] T.Geng*, R.Shi*, P.Dong*, ..., M.Herbordt, A.Li, Y.Wang: "CSB-RNN: A Faster-than-Realtime RNN Acceleration Framework with Compressed Structured Blocks", the 34th ACM International Conference on Supercomputing.

   2019:

  • [ICS 2019] T.Geng, T.Wang, C.Wu, C.Yang, W.Wu, A.Li, M.Herbordt: "O3BNN: An Out-Of-Order Architecture for High-Performance Binarized Neural Network Inference with Fine-Grained Pruning", the 33th ACM International Conference on Supercomputing.

  • [SC 2019] A.Li, T.Geng, T.Wang, M.Herbordt, S.Song, K.Barker: "BSTC: A Novel BinarizedSoft-Tensor-Core Design for Accelerating Bit-Based Approximated Neural Nets", Proceedings of the International Conference for High-Performance Computing, Networking, Storage, and Analysis.

  • [SC 2019] C.Yang, T.Geng, T.Wang, ..., M.Herbordt: "Fully integrated FPGA molecular dynamics simulations", Proceedings of the International Conference for High-Performance Computing, Networking, Storage, and Analysis.

publicaton
Projects

Projects

Screenshot 2024-10-02 231234.png
Screenshot 2024-10-02 230942.png
Screenshot 2024-10-02 225857.png

Sustainable Generative AI

Pioneering next-generation GenAI through innovative (digital)architecture-algorithm co-design, we aim to jointly enhance efficiency, creativity, and trustworthiness in GenAI inference, pretraining, and fine-tuning, forging a path toward sustainable AI development. Our recent interests primarily focus on LLMs and Diffusion Models.

AI for Broadly-Defined Science

Recognizing the soaring expressivity of GenAI, we are pioneering its application in scientific exploration, with a recent focus on nuclear fusion, smart grids, financial networks, and pandemic control. Our current interests include LLMs with prompt tuning and domain-specific diffusion models for complex system data generation

Nature-Powred AI Profecessor

‘Using Nature as a Low-Power Supercomputer’ is no longer a dream! We’ve developed new AI processors that harness nature's computational power to handle diverse AI workloads, such as Graph Learning, LLM inference, pretraining, and fine-tuning. Nature offers a 1000x speedup and a 100,000x cost reduction over GPUs

Team

Meet The Team

chunshu_edited.jpg

Chunshu Wu 

Research Interests:

1. Dynamical System for ML;

2. Computer Architecture;

3. AI for Science;

WeChat Image_20221216175330.jpg

Clein Song

Research Interests:

1. Computer Architecture

2. Dynamical-System-based Compute Substrate

3. Generative AI Codesign

WeChat Image_20230130115704.jpg

Chuan Liu

Research Interests:

1. Graph Learning

2. Generative AI for Science

3. Dynamical-System-based Machine Learning

1528805090361.jpg

Ali Falahati

Research Interests:

1. Efficient Generative AI

2. GPU Optimization

3. Large Language Model

News
Awards & Services

News

09/2024 One paper accepted by NeurIPS 2024.

09/2024 Prof. Tony Geng received Research Fund from DOE OS on AI4Science.

08/2024 Ali Falahati joined the big family. Welcome!

08/2024 Prof. Tony Geng received Research Fund from the New York State Center of Excellence in Data Science on Next-Generation Sustainable AI Powered by Dynamical Systems.

08/2024 Prof. Tony Geng received Research Fund from DOE OE on GNN for Smart Grid Management.

08/2024 One paper accepted by Briefings in Bioinformatics 2024 - Impact Factor: 13.99.

08/2024 One paper accepted by Advanced Intelligent Systems 2024 - Impact Factor: 7.4.

07/2024 One paper accepted by MICRO 2024.

06/2024 One paper accepted by SC 2024. 

06/2024 Prof. Tony Geng received Research Award from NERSC on GenAI -- providing 40000 GPU hours. 

06/2024 Prof. Tony Geng received Research Fund from PNNL/DOE on Mixed-Precision Generative AI.

05/2024 Pouya will join Meta in August as an AI chip architect with an unbelievable package. Big Congrats to Pouya!!! 

05/2024 Prof. Tony Geng received the Award for Excellence in Graduate Teaching!

05/2024 One paper accepted by ICML 2024. 

04/2024 One paper accepted by ICS 2024. 

03/2024 One paper accepted by ISCA 2024. 

01/2024 One paper accepted by TC 2024

01/2024 One paper accepted by ICLR 2024: Extending binary Ising machines to real-valued dynamical systems through hardware architecture and Hamiltonian codesign for graph learning problems. 

10/2023 Prof. Tony Geng received Research Fund from META Reality Lab on Next-generation Graphics Architecture.

09/2023 Two papers accepted by NeurIPS 2023.

08/2023 Prof. Tony Geng received Research Fund from NSF CORE program on GNN Acceleration based on Digital Hardware.

07/2023 Prof. Tony Geng received Research Fund from the New York State Center of Excellence in Data Science on Nature-powered Machine Learning.

07/2023 One paper accepted by MICRO 2023.

07/2023 One paper accepted by ICCAD 2023.

07/2023 One paper accepted by ICCV 2023.

07/2023 One paper accepted by JPCC (Journal of Physical Chemistry C) 2023.

06/2023 One paper accepted by SC 2023.

06/2023 Prof. Tony Geng received Donations from AMD and Xilinx, thanks!

04/2023 One paper accepted by IEEE Transactions on Image Processing (TIP) 2023 - Impact Factor: 10.86.

04/2023 Two papers accepted by ICS 2023 -- SmartNIC and SmartSwitch can significantly improve DLRM and GNN training efficiency.

03/2023 One paper accepted by OSDI 2023.

03/2023 Prof. Tony Geng received Research Fund from PNNL/DOE on Efficient Data Format for Large Language Model.

02/2023 One paper accepted by CVPR 2023 (as a Highlighted Paper).

02/2023 Three papers accepted by DAC 2023.

11/2022 One paper accepted by AAAI 2023.

10/2022 One paper accepted by TPDS 2022.

09/2022 Prof. Tony Geng received Faculty Research Award from META (Facebook) on AI System Hardware/Software Codesign.

09/2022 Our proposal was selected as an internationally excellent finalist in Meta (Facebook) RFP - Networking for AI.

09/2022 Four papers were accepted by ICCD 2022.

06/2022 Three papers were accepted by FPL 2022.

04/2022 Two papers were accepted by ICS 2022.

02/2022 One paper was accepted by DAC 2022.

12/2021 Two papers were accepted by HPCA 2022.

Alumni

Pouya Haghi -- 2024.8: Joined META as AI chip architect

Sponsors

download.jpg
download (2).png
download (3).png
meta-logo.png
download (1).png
download (4).png
Alumni
bottom of page