Bio

Welcome to my website! I am a 5th year PhD candidate in Computer Science at Johns Hopkins University. My research lies at the intersection of systems and machine learning. Specifically, I look at I/O and data movement bottlenecks in machine learning algorithms. I am fortunate to be advised by Dr Randal Burns and Dr Joshua Vogelstein. During my PhD I spent summers at Meta (2021) and Argonne National Laboratory (2022). Prior to joining JHU, I earned my BS and MS degrees at the International Institute of Information Technology, Bangalore. Contact me at @jhu.edu for questions about research or potential collobarations.

Pre-prints

1. B. Wheatman, M. Madhyastha and R. Burns. Masked Matrix Multiplication for Emergent Sparsity. Arxiv preprint [PDF]

Publications

1. R. Underwood, M. Madhyastha, R. Burns, B. Nicolae. EvoStore: Towards Scalable Storage of Evolving Learning Models. Proceedings of International Conference on Supercomputing (HPDC 2024)[PDF]
2. M. Madhyastha, T. Budavari, V. Braverman, J. Vogelstein and R. Burns. T-Rex (Tree-Rectangles): Reformulating Decision Tree Traversal as Hyperrectangle Enclosure. Proceedings of International Conference on Data Engineering (ICDE 2024) [PDF]
3. M. Madhyastha, R. Underwood, R. Burns, B. Nicolae. A Lightweight Scalable Learning Model Repository with Fine-Grain Tensor-Level Access. Proceedings of International Conference on Supercomputing (ICS 2023)[PDF]
4. M. Madhyastha, K. Lillaney, J. Browne, J. Vogelstein and R. Burns. BLOCKSET(Block-Aligned Serialized Trees): Reducing Inference Latency for Tree-Ensemble Deployment.Proceedings of the Proceedings of The 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining.[KDD 2021][PDF]
5. M. Madhyastha, G. Li, V. Strnadova-Neeley, J. Browne, J. Vogelstein, R. Burns and C. Priebe. Geodesic Forests.Proceedings of the Proceedings of The 26th ACM SIGKDD Conference on Knowledge Discovery and Data Mining.[KDD 2020][PDF]
6. S. Roy, M. Madhyastha, S. Lawrence and V. Ranjan. Deep Learning based Method to Infer Concept Prerequisite Relations from Online Educational Resources. [IAAI 2018(co-located with AAAI-2018)[PDF]
7. M. Madhyastha, S. Reddy and S. Rao. Online Scheduling of a Fleet of Autonomous Vehicles Using Agent-Based Procurement Auctions. Proceedings of the 2017 IEEE International Conference on Service Operations and Logistics and Informatics (SOLI), IEEE Intelligent Transportation Systems Society. Bari, Italy [PDF]
8. M. Madhyastha and D. Jayagopi. A Low Cost Personalised Robot Language Tutor with Perceptual and Interaction Capabilities. Proceedings of the 13th IEEE Annual India Conference (INDICON), 2016. Bangalore, India. [PDF]

Professional Experience

Summer 2022: Research intern at Argonne National Laboratory Summer 2021: Machine learning intern at Meta Summer 2018: Intern at CERN Summer 2017: Research Intern at INRIA [Parietal Lab] Summer 2015: Research Intern at Siemens R&D

Teaching

Parallel Programming (JHU CS320/620): Teaching Assistant Spring 2020, Fall 2020, Fall 2021, Fall 2022

Miscellaneous/Hobbies

I’m also an open source enthusiast and I’ve contributed to various projects such as Boost, SymPy and the linux kernel. Outside of research, I’m an avid outdoor enthusiast. I enjoy hiking, biking and nature walks. I’m also trained hindustani classical vocalist and obtained my junior diploma in music from Akhil Bharatiya Gandharva Mahavidyalaya Mandala. I love travelling and cooking.