Scholar
Balasubramaniam Srinivasan
Google Scholar ID: uM4EhgEAAAAJ
Amazon (AWS)
Citations & Impact
Citations
1,403
H-index
11
i10-index
12
Publications
17
Co-authors
0
Publications
Resume (English only)
Academic Achievements
- Recent publications are mentioned but specific titles or details were not provided.
Research Experience
- Graduate Research Assistant at Purdue University since August 2018, focusing on graph representation learning, equivariant and invariant representation learning.
- Applied Scientist Intern at Amazon from May 2021 to August 2021, proposed a message passing neural network to capture non-rigidity of protein molecules, defining conditional transformations via conditional group equivariances and invariances.
- Applied Scientist Intern at Amazon from June 2020 to September 2020, proposed a hypergraph neural network that exploited the incidence structure, providing provably expressive representations of vertices, hyperedges, and the complete hypergraph, introducing a new task on hypergraphs - variable-sized hyperedge expansion.
- Software Engineer Intern at Salesforce from June 2017 to September 2017, performed anomaly detection on page load time data using Spark for incremental spectral clustering and root cause analysis.
- Software Engineer at ARM from July 2015 to July 2016, analyzed SPEC and streaming workload performance for mobile and enterprise systems, developing benchmarks for cache hierarchy and memory controllers, and created an architecture-agnostic power model.
- Research Intern at Indian Institute of Science from January 2015 to May 2015, designed a runtime resource manager for a massively parallel dynamically reconfigurable accelerator, developing kernel modules and a host user application for device driver support.
Education
- PhD in Computer Science, 2018-2022, Purdue University; M.S. in Computer Science, 2016-2018, UC San Diego; B.E. (Hons) in Electrical and Electronics Engineering, 2011-2015, BITS Pilani.
Background
- Balasubramaniam (Bala) is a fourth-year PhD student in the Computer Science department at Purdue University. His research interests lie in applying group theory, representation theory, and invariant theory to deep learning, enriching neural networks with knowledge about the structures and symmetries in data. His work finds real-world applications in sets, images, graphs (e.g., social networks, molecules, etc.).
Miscellany
- No information provided regarding personal interests or hobbies.
Co-authors