Its Me

Don Dennis

Ph.D. Student
Machine Learning Department
Carnegie Mellon University
  donkdennis [at] gmail [dot] com
  dondennis [at] cmu [dot] edu
  Pittsburgh, PA, USA
  Google Scholar
  Github
  Resume

I am a final year PhD student in the Machine Learning Department at Carnegie Mellon University, advised by Prof. Virginia Smith. My work spans the intersection of machine learning (ML), optimization, and systems.

I aim to understand, design, and implement ML systems that are both theoretically sound and practically impactful, bridging low-level hardware, algorithms, and high-level applications. My work includes developing algorithms and frameworks to

  • Perform on-device ML inference with minimal memory and compute (e.g., EdgeML on microcontrollers),
  • Enable efficient federated learning on heterogeneous networks (e.g., k-fed: one-shot federated clustering),
  • Support adaptive inference for cloud-based LLM deployments (e.g., B-Distil, ABC),
  • Continually solve large-scale optimization problems for resource allocation (e.g., Deep Learning cluster scheduling, Network Traffic Engineering).

In Summer '21, I interned with Kazuhito Koishida at Microsoft Applied Sciences Group, Redmond, where I worked on adaptive ML inference methods. Before graduate school, I spent two enriching years as a Research Fellow at Microsoft Research India, collaborating with Prateek Jain and Harsha Simhadri. I completed my undergraduate studies at the Indian Institute of Technology, Patna, where I worked with Prof. Arijit Mondal on designing a RISC-V microprocessor and accompanying toolchain, applied in FPGA-based hardware simulations and educational tools.


   I am on the current job market (2025) and am looking for opportunities as a researcher in machine learning and systems, where I can contribute to advancing scalable, resource-efficient ML systems.

Publications

Agreement-Based Cascading for Efficient Inference
Don Dennis*, Steven Kolawole*, Ameet Talwalkar, Virginia Smith
In submission at Conference on Machine Learning and Systems (MLSys), 2025
arXiv preprint arXiv:2407.02348
Progressive Ensemble Distillation: Building Ensembles for Efficient Inference
Don Dennis, Abhishek Shetty, Anish Sevekari, Kazuhito Koishida, Virginia Smith
Advances in Neural Information Processing Systems (NeurIPS), 2023.
Bitrate-Constrained DRO: Beyond Worst Case Robustness To Unknown Group Shifts
Amrith Setlur, Don Dennis, Benjamin Eysenbach, Aditi Raghunathan, Chelsea Finn, Virginia Smith and Sergey Levine
International Conference on Learning Representations (ICLR), 2023.
Heterogeneity for the Win: Communication Efficient Distributed Clustering
Don Dennis, Tian Li, Virginia Smith
International Conference on Machine Learning (ICML), 2021 (Spotlight).
Shallow RNN: Accurate Time-series Classification on Resource Constrained Devices
Don Dennis, Durmus Alp Emre Acar, Vikram Mandikal, Vinu Sankar Sadasivan, Harsha Vardhan Simhadri, Venkatesh Saligrama, Prateek Jain
Advances in Neural Information Processing Systems (NeurIPS), 2019
GesturePod: Programmable Gesture Recognition for Augmenting Assistive Devices
Shishir Patil, Don Dennis, Chirag Pabbaraju, Rajanikant Deshmukh, Harsha Vardhan Simhadri, Manik Varma, Prateek Jain
ACM User Interface Software and Technology Symposium (UIST), 2019
Multiple Instance Learning for Sequential Data Classification on Resource Constrained Devices
Don Dennis, Chirag Pabbaraju, Harsha Vardhan Simhadri, Prateek Jain
Advances in Neural Information Processing Systems (NeurIPS), 2018
Single Cycle RISC-V Micro Architecture Processor and its FPGA Prototype
Don Dennis, Ayushi Priyam, Sukhpreet Singh Virk, Sajal Agrawal, Tanuj Sharma, Arijit Mondal, Kailash Chandra Ray
International Symposium on Embedded Computing and System Design (ISED), 2017

Workshops and Demos

Revisiting Cascaded Ensembles for Efficient Inference
Steven Kolawole, Don Dennis, Ameet Talwalkar, Virginia Smith
ES-FOMO Workshop at International Conference on Learning Representations (ICLR), 2024
Progressive Knowledge Distillation: Balancing Inference Latency and Accuracy at Runtime
Don Dennis, Abhishek Shetty, Anish Sevekari, Kazuhito Koishida, Virginia Smith
ES-FOMO Workshop at International Conference on Machine Learning (ICML), 2024
EdgeML: Edge of Machine Learning --- Demonstration of Low Resource Keyword Spotting
Don Dennis, Harsha Simhadri, Prateek Jain
MLPCD2 Workshop at Advances in Neural Information Processing Systems (NeurIPS), 2018
GesturePod: Demonstrating On-Device Gesture Recognition
Shishir Patil, Don Dennis, Chirag Pabbaraju, Harsha Simhadri, Manik Varma, Prateek Jain
Microsoft Booth 203, Advances in Neural Information Processing Systems (NeurIPS), 2018
Talk-Bot: Federated Human Detection for Collaborative Multi-Angle Videography
Don Dennis, Harshit Singh, Karan Jakhar, Prashant Baghel
International Symposium on Embedded Computing and System Design (ISED), 2016