SUNNYVALE, CA, December 17, 2018 – ParallelM Senior Architect Swaminathan (Swami) Sundararaman will present the paper, “Interpretability and Reproducibility in Production Machine Learning Applications,” at the 17th Annual IEEE International Conference on Machine Learning and Applications (ICMLA) in Plaza D at the Orlando Hyatt Regency on December 18th at 8:30 a.m. Mr. Sundararaman will discuss techniques using canary models and how the capture of model behavior timelines can be used to understand model behavior in production environments.
WHAT: “Interpretability and Reproducibility in Production Machine Learning Applications,” IEEE International Conference on Machine Learning and Applications
ParallelM Lead Architect Swami Sundararaman will discuss the complexities of machine learning interpretability and reproducibility in production environments. The presentation will discuss why Explainability/Interpretability in machine-learning (ML) applications is becoming critical, with legal and industry requirements demanding human understandable machine-learning results. Mr. Sundararaman will describe the additional complexities that occur when a known interpretability technique (canary models) is applied to a real production scenario. He will furthermore argue that reproducibility is a key feature in practical usages of such interpretability techniques in production scenarios. With this motivation, he will present a production ML reproducibility solution, namely a comprehensive time ordered event sequence for machine-learning applications. Mr. Sundararaman will demonstrate how this approach can bring this known common interpretability technique into production viability. He will further present the system design and early performance characteristics of ParallelM’s reproducibility solution.
WHO: Swaminathan (Swami) Sundararaman is the Lead Architect of ParallelM, a startup focused on production machine learning and deep learning. Mr. Sundararaman was previously at Fusion-io Inc. & Sandisk Corp., where he focused on innovations in non-volatile memory technologies and applications. He holds a PhD from the University of Wisconsin-Madison, Madison, MS from Stony Brook University, New York, and a Bachelor of Computer Science from BITS, Pilani, India. Swami holds 20 patents in operating systems, non-volatile memory, and storage systems. He received best paper awards at Eurosys’14, FAST ’10, and USENIX ATC ’09 for his research work. His research interests include distributed systems, machine learning, operating systems, and file & storage systems.
WHEN: December 18th at 8:30am
WHERE: 17th Annual IEEE International Conference on Machine Learning and Applications, Plaza D at the Hyatt Regency in Orlando Florida
ParallelM is the first and only company completely focused on delivering machine learning operationalization (MLOps) at scale. ParallelM’s breakthrough MCenter™ solution is built specifically to power the deployment, optimization, and governance of machine learning pipelines in production so that companies can scale machine learning across their business applications. ParallelM’s approach is that of a single, unified MLOps solution that embeds best practice processes in technology, enabling all ML stakeholders to unlock the business value of AI. Please visit www.parallelm.com or email us at firstname.lastname@example.org.
Marianne Dempsey/Jenna Beaucage
Phone: 508-475-0025 x115/ x124