TechFest Workshop – Theory Day – Session 3
- Ravishankar Krishnaswamy | MSR India
We present the first non-trivial online algorithms for the classical multi-commodity buy-at-bulk problem. The goal is to provision a low-cost network over time to support bandwidth demands that also arrive over time. Previously, online algorithms were known only for the special cases of all links having the same cost (Awerbuch and Azar, FOCS 1997), or when all demands share the same destination (Meyerson, SPAA 2004).. At the crux of our algorithm is a generic online reduction for a class of network design problems from multi-commodity demands to single-commodity instances. (Joint work with Deeparnab Chakrabarty, Alina Ene and Debmalya Panigrahi.)
-
-
Ben Ryon
-
Ravishankar Krishnaswamy
Principal Researcher
-
-
Watch Next
-
-
Fuzzy Extractors are Practical
- Melissa Chase,
- Amey Shukla
-
-
-
Microsoft Research India - The lab culture
- P. Anandan,
- Indrani Medhi Thies,
- B. Ashok
-
GenAI for Supply Chain Management: Present and Future
- Georg Glantschnig,
- Beibin Li,
- Konstantina Mellou
-
Using Optimization and LLMs to Enhance Cloud Supply Chain Operations
- Beibin Li,
- Konstantina Mellou,
- Ishai Menache
-
-
-