Hadi S. Esmaeilzadeh
Halicioğlu Chair in Computer Architecture
Associate Professor
University of California, San Diego
Department of Computer Science and Engineering
hadi [AT] eng [DOT] ucsd [DOT] edu
9500 Gilman Drive, #0404
La Jolla, CA 92093-0404
Research Group: Alternative Computing Technologies (ACT) Laboratory
Curriculum Vitae
Hadi was awarded early tenure at the University of California, San Diego (UCSD), where he is an associate professor in Computer Science and Engineering. Prior to UCSD, he was an assistant professor in the School of Computer Science at the Georgia Institute of Technology from 2013 to 2017. There, he was the inaugural holder of the Catherine M. and James E. Allchin Early Career Professorship. Hadi is the founding director of the Alternative Computing Technologies (ACT) Lab, where his team is developing new technologies and cross-stack solutions to build the next generation computer systems. He is also the associate director of Center for Machine Integrated Computing and Security (MICS) at UCSD.
Dr. Esmaeilzadeh received his Ph.D. from the Department of Computer Science and Engineering at the University of Washington in 2013. His Ph.D. dissertation received the 2013 William Chan Memorial Dissertation Award from the University of Washington. Hadi was inducted to the ISCA Hall of Fame in 2018.
His research has been recognized by four Communications of the ACM Research Highlights, four IEEE Micro Top Picks, recently one more nominations for Communications of the ACM Research Highlights, one more honorable mention in IEEE Micro Top Picks, and a Distinguished Paper Award in HPCA 2016.
Hadi has received the Air Force Young Investigator Award (2017), College of Computing Outstanding Junior Faculty Research Award (2017), Qualcomm Research Award (2017 and 2016), Google Research Faculty Award (2016 and 2014), Microsoft Research Award (2017 and 2016), and Lockheed Inspirational Young Faculty Award (2016).
His team was awarded the Qualcomm Innovation Fellowship (2014), one of his student is a Microsft Research Fellow, and another won the the 2017 National Center for Women and IT (NCWIT) Collegiate Award. Four of his undergradaute students have been awarded the Georgia Tech President's Undergraduate Research Award (PURA). Hadi's work on dark silicon has been profiled in New York Times.
His research has been recognized by four Communications of the ACM Research Highlights, four IEEE Micro Top Picks, recently one more nominations for Communications of the ACM Research Highlights, one more honorable mention in IEEE Micro Top Picks, and a Distinguished Paper Award in HPCA 2016.
Hadi has received the Air Force Young Investigator Award (2017), College of Computing Outstanding Junior Faculty Research Award (2017), Qualcomm Research Award (2017 and 2016), Google Research Faculty Award (2016 and 2014), Microsoft Research Award (2017 and 2016), and Lockheed Inspirational Young Faculty Award (2016).
His team was awarded the Qualcomm Innovation Fellowship (2014), one of his student is a Microsft Research Fellow, and another won the the 2017 National Center for Women and IT (NCWIT) Collegiate Award. Four of his undergradaute students have been awarded the Georgia Tech President's Undergraduate Research Award (PURA). Hadi's work on dark silicon has been profiled in New York Times.
I am eagerly looking for students that ambitiously want to make a difference! Please read my research statement before contacting me.
Project PHI: System Design for Pervasive Hierarchal Intelligence
Currently,
we are focusing on Project PHI (Pervasive Hierarchical
Intelligence), a holistic effort to provide a comprehensive
solution for making immersive machine intelligence a
reality. Our guiding principle is to retain as much
generality and automation while delivering large performance
and efficiency gains through specialization and acceleration
for a wide range of learning and intelligence workloads. As
the first milestone of Project PHI, we have developed Tabla,
which is open source and available at http://act-lab.org/artifacts/tabla/.
This cross-stack solution - spanning from programming
language to the hardware - rethinks the hardware/software
abstraction by delving into the theory of machine learning.
It leverages the insight that many learning algorithms can
be solved using stochastic gradient descent that minimizes
an objective function. The solver is fixed while the
objective function changes with the learning algorithm. Therefore,
Tabla
uses stochastic optimization as the abstraction between
hardware and software. Consequently, programmers specify the
learning algorithm by merely expressing the gradient of the
objective function in our domain specific language. Tabla
then automatically generates the synthesizable
implementation of the accelerator for scale-out FPGA
realization using a set of template designs. Real hardware
measurements show orders of magnitude higher performance and
power efficiency while the programmer only writes 60 lines
of code. These encouraging results show that rethinking the
hardware/software abstractions from an algorithmic
perspective can open new dimensions in system design for
Pervasive Hierarchical Intelligence.
Teaching
- CSE 141: Introduction to Computer Architecture : In Winter 2019, I taught The Introduction to Computer Architecture course.
- CSE 240D: Accelerator Design for Deep Learning: In Winter 2019, I am taught the graduate Accelerator Design for Deep Learning course.
- CSE 240A: Principles of Computer Architecture: In Fall 2018, I taught the graduate Computer Architecture cource.
- CSE 240C: Accelerated Architecture Design for Machine Learning and Artificial Intelligence: In Winter 2018, I taught the Advanced Microarchitecture cource.
- CS 3220: Processor Design with FPGAs: In Fall 2017, I taught the Processor Design course.
- CS 8803 ACT: Alternative Computing Technologies: In Spring 2017, I taught the Alternative Computing Technologies course.
- CS 3220: Processor Design with FPGAs: In Fall 2016, I taught the Processor Design course.
- CS 8803 ACT: Alternative Computing Technologies: In Spring 2016, I taught the Alternative Computing Technologies course.
- CS 3220: Processor Design with FPGAs: In Fall 2015, I taught the Processor Design course.
- CS 4290/CS 6290/ECE 4100/ECE 6100 HPCA: In Spring 2015, I taught the High Performance Computer Architecture course.
- CS 3220: Processor Design with FPGAs: In Fall 2014, I taught the Processor Design course.
- CS 8803 ACT: Alternative Computing Technologies: In Spring 2014, I taught a new course on alternative computing technologies.
PhD Students
- Shu-Ting Wang
- Parsa Assadi
- Hanyang Xu
- Soroush Ghodrati
- Byung Hoon Ahn
- Sean Kinzer
- Rohan Mahapatra
MS Students
- Edwin Mascarenhas
Undergraduate Students
- Michael Brzozowski (Winner of Georgia Tech President's Undergraduate Research Award (PURA) for Fall 2018)
- Preston Olds (Winner of Georgia Tech President's
Undergraduate Research Award (PURA) for Fall 2017)
- Rafael Oliviera
- Austin R Leal
- Xiaoyi Cai (Winner of Georgia Tech President's
Undergraduate Research Award (PURA) for Fall 2017)
Past Undergraduate Students
- Chenkai Shao (Winner of Georgia Tech President's
Undergraduate Research Award (PURA) for Spring 2016)
- Graham Miles
- Elizabeth Perakis
Alumni
- Joon Kyung Kim
- Ahmed Taha Elthakeb, CSE, PhD. First Employment: Samsung, San Diego, CA
- Hardik Sharma, CS, PhD. First Employment: Bigstream, Mountain View, CA
- Divya Mahajan, CS, PhD. First Employment: Microsoft, Redmond, WA
- Jongse Park, CS, PhD. First Employment: Bigstream, Mountain View, CA
- Amir Yazdanbakhsh, CS, PhD. First Employment: Google Brain Residency, Mountain View, CA
- Brahmendra Reddy Yatham, CSE, MS. First Employment: NVIDIA, Santa Clara, CA
- Prannoy Pilligundla, CSE, MS. First Employment: Amazon AI, Palo Alto, CA
- Jacob I. Sacks, CS, PhD. First Employment: PhD Program School of Interactive Computing, Georgia Institute of Technology, Atlanta, GA
- Emmanuel
Amaro, CS, M.Sc., First Employment: PhD Program,
University of California, Berkeley, Berkeley, CA
- Bradley
Thwaites, ECE, M.Sc., First Employment: ARM, Austin,
TX
- Sindhuja Sethuraman, ECE, M.Sc., First Employment: Apple Inc., Orlando, FL
- Anandhavel Nagendrakumar, ECE, M.Sc., First Employment: IM Flash Technologies, Lehi, UT
- Physically Accurate Learning-based Performance Prediction of Hardware-accelerated ML Algorithms, Workshop on Machine Learning for CAD (MLCAD), 2022
- Yin-Yang: Programming Abstractions for 4 Cross-Domain Multi-Acceleration, IEEE Micro, Special Issue on Compiling for Accelerators, 2022
- Glimpse: Mathematical Embedding of Hardware Specification for Neural Compilation, DAC, 2022 (Nominated for the Best Paper Award)
- Conscious Empathic AI in Service, Journal of Service Research, 2022
- VeriGOOD-ML: An Open-Source Flow for Automated ML Hardware Synthesis, International Conference On Computer Aided Design (ICCAD), 2021
- Not All Features Are Equal: Discovering Essential Features for Preserving Prediction Privacy, Web Conference (WWW), 2021
- A Computational Stack for Cross-Domain Acceleration, HPCA, 2021 (Artifact)
- Planaria: Dynamic Architecture Fission for Spatial Multi-Tenant Acceleration of Deep Neural Networks, MICRO, 2020
- Ordering Chaos: Memory-Aware Scheduling of Irregularly Wired Neural Networks for Edge Devices, MLSys, 2020
- SinReQ: Generalized Sinusoidal Regularization for Low-Bitwidth Deep Quantized Training, ICML Workshop on Understanding and Improving Generalization in Deep Learning, 2019
- Divide and Conquer: Leveraging Intermediate Feature Representations for Quantized Training of Neural Networks, ICML Workshop on Understanding and Improving Generalization in Deep Learning, 2019
- Shredder: Learning Noise to Protect Privacy with Partial DNN Inference on the Edge, ICML Security and Privacy of Machine Learning workshop, 2019
- RELEASE: Reinforcement Learning and Adaptive Sampling for Optimized DNN Compilation, ICML RL for Real Life workshop, 2019
- ReLeQ: A Reinforcement Learning Approach for Deep Quantization of Neural Networks, NeurIPS ML for Systems workshop, 2018
- From Tensors to FPGAs: Accelerating Deep Learning, Hot Chips, 2018
- A Network-Centric Hardware/Algorithm Co-Design to Accelerate Distributed Training of Deep Neural Networks, MICRO, 2018
- In-RDBMS Hardware Acceleration of Advanced Analytics , VLDB, 2018
- RoboX: An End-to-End Solution to Accelerate Autonomous Control in Robotics , ISCA, 2018 (lightning talk)
- Bit Fusion: Bit-Level Dynamically Composable Architecture for Accelerating Deep Neural Networks , ISCA, 2018 (lightning talk) (Artifact - Coming soon)
- GANAX: A Unified MIMD-SIMD Acceleration for Generative Adversarial Networks , ISCA, 2018 (lightning talk)
- SnaPEA: Predictive Early Activation for Reducing Computation in Deep Convolutional Neural Networks , ISCA, 2018 (lightning talk)
- SiMul: An Algorithm-Driven Approximate Multiplier Design for Machine Learning, IEEE Micro, special issue on Approximate Computing, 2018
- FlexiGAN: An End-to-End Solution for FPGA Acceleration of Generative Adversarial Networks , FCCM, 2018
- A SIMD-MIMD Acceleration with Access-Execute Decoupling for Generative Adversarial Networks , SysML, 2018
- Scale-Out Acceleration for Machine Learning, MICRO, 2017
- Proving Flow Security of Sequential Logic via Automatically-Synthesized Relational Invariants, CSF, 2017
- AxBench: A Multiplatform Benchmark Suite for Approximate Computing, IEEE Design & Test, 2017 (Artifact)
- The Impact of 3D Stacking on GPU-Accelerated Deep Neural Networks: an Experimental Study, 3DIC, 2016
- From High-Level Deep Neural Models to FPGAs, MICRO, 2016 (Artifact)
- Towards Statistical Guarantees in Controlling Quality Tradeoffs for Approximate Acceleration, ISCA, 2016
- AxGames: Towards Crowdsourcing Quality Target Determination in Approximate Computing, ASPLOS, 2016 (Artifact)
- DNNWEAVER: From High-Level Deep Network Models to FPGA Acceleration Workshop on Cognitive Architectures (CogArch) in conjunction with ASPLOS, 2016.
- TABLA: A Unified Template-Based Framework for Accelerating Statistical Machine Learning, HPCA, 2016 (Distinguished Paper Award) (Artifact)
- GRATER: An Approximation Workflow for Exploiting Data-Level Parallelism in FPGA Acceleration, DATE, 2016
- RFVP: Rollback-Free Value Prediction with Safe-to-Approximate Loads, ACM TACO and HiPEAC, 2015
- TABLA: A Unified Template-based Framework for Accelerating Statistical Machine Learning, SCS Technical Report, GT-CS-15-07, Georgia Institute of Technology, September, 2015
- Mitigating the Memory Bottleneck with Approximate Load Value Prediction, IEEE Design & Test, 2015
- Neural Acceleration for GPU Throughput Processors, MICRO, 2015
- Axilog: Abstractions for Approximate Hardware Design and Reuse, IEEE Micro, special issue on Alternative Computing Designs and Technologies, 2015
- Prediction-Based Quality Control for Approximate Accelerators, Workshop on Approximate Computing Across the System Stack (WACAS) in conjunction with ASPLOS, 2015
- ExpAX: A Framework for Automating Approximate Programming, SCS Technical Report, GT-CS-14-05, Georgia Institute of Technology, July, 2014
- Rollback-Free Value Prediction for Approximate Loads, PACT, 2014
- General-Purpose Code Acceleration with Limited-Precision Analog Computation, ISCA, 2014 (Honorable Mention, IEEE Micro Top Picks)
- A Reconfigurable Fabric for Accelerating Large-Scale Datacenter Services, ISCA, 2014 (selected for Communications of the ACM Research Highlights and IEEE Micro Top Picks)
- Methodical Approximate Hardware Design and Reuse, Workshop on Approximate Computing Across the System Stack (WACAS) in conjunction with ASPLOS, 2014
- Toward General-Purpose Code Acceleration with Analog Computation, Workshop on Approximate Computing Across the System Stack (WACAS) in conjunction with ASPLOS, 2014
- Expectation-Oriented Framework for Automating Approximate Programming, Workshop on Approximate Computing Across the System Stack (WACAS) in conjunction with ASPLOS, 2014
- Bio-Accelerators: Bridging Biology and Silicon for General-Purpose Computing, Wild and Crazy Ideas (WACI) in conjunction with ASPLOS, 2014
- Expectation-Oriented Framework for Automating Approximate Programming, SCS Technical Report, GT-CS-13-07, Georgia Institute of Technology, December, 2013
- Neural Acceleration for General-Purpose Approximate Programs, IEEE Micro Top Picks, 2013 (original in MICRO '12)
- Power Challenges May End the Multicore Era, Communication of the ACM (CACM) Research Highlights, 2013 (Pradip Bose's perspective on our paper)
- How to Implement Effective Prediction and Forwarding for Fusable Dynamic Multicore Architectures, HPCA, 2013
- Neural Acceleration for General-Purpose Approximate Programs, MICRO, 2012 (selected for Communications of the ACM Research Highlights and IEEE Micro Top Picks)
- Multicore Model from Abstract Single Core Inputs, Computer Architecture Letters (CAL), 2012
- Power Limitations and Dark Silicon Challenge the Future of Multicore, ACM Transaction on Computer Systems (TOCS), 2012 (invited, original in ISCA '11)
- Looking Back and Looking Forward: Power, Performance, and Upheaval, Communications of the ACM (CACM) Research Highlights, 2012 (David Patterson's perspective on our paper)
- Addressing Dark Silicon Challenges with Disciplined Approximate Computing, Dark Silicon (DaSi) Workshop, 2012
- Toward Neural Acceleration for General-Purpose Approximate Programs, Workshop on Energy-Efficient Design (WEED), 2012
- Dark Silicon and the End of Multicore Scaling, IEEE Micro Top Picks 2012 (original in ISCA '11)
- What Is Happening to Power, Performance, and Software?, IEEE Micro Top Picks from the 2011 Computer Architecture Conferences, 2012 (original in ASPLOS '11)
- Architecture Support for Disciplined Approximate Programming, ASPLOS, 2012
- Dark Silicon and the End of Multicore Scaling, ISCA, 2011 (selected for Communications of the ACM Research Highlights and IEEE Micro Top Picks)
- Looking
Back on the Language and Hardware Revolution: Measured
Power, Performance, and Scaling, ASPLOS, 2011 (Measurement
Data) (selected for Communications of the ACM
Research Highlights and IEEE Micro Top Picks)
- Tonbak (Persian goblet drum)
- Daf (chained Persian frame drum)
Recent Publications:
For a full list, please check out my Google Scholar profile!
Music:
Guest Performer. Concert in Dashti. Bereket UT-Austin
Middle Eastern Ensemble, Butler School of Music, The
University of Texas at Austin, Bates Concert Hall, April 2010