Athresh Karanam

PhD candidate at UTD CS | ML/generative models/optimization | BTech: IITG

[Saurabh Mathur @mathursaurabh96]

In our recent #AAAI25 work, we propose a unified framework for expressing domain knowledge about variables like monotonicity & independence. We use this framework to learn complex Probabilistic Circuit models from small & noisy data sets by exploiting multiple forms of knowledge.

[Athresh Karanam @athresh55]

Glad I had the opportunity to present our work on explaining sum-product networks using a tree-structured representation of context-specific independencies encoded by the SPN. Thanks to @mathursaurabh96, Dr. David Haas, @pedjagogue, @kerstingAIML and my advisor @Sriraam_UTD.

[PGM'2022 Almería @pgm2022almeria]

Bhagirath Athresh Karanam, Saurabh Mathur, Predrag Radivojac, David M Haas, Kristian Kersting and Sriraam Natarajan. Explaining Deep Tractable Probabilistic Models: The sum-product network case. #PGM2022

[Kristian Kersting @kerstingAIML]

🚨🚨🚨new #PSB23 paper alert. We model gestational diabetes in a clinical study & develop a domain expert-guided probabilistic model with causal independence (Noisy-Or) pdf 👉 https://t.co/mw5MdWk2NV 🙏 @athresh55 @mathursaurabh96 @pedjagogue @DMHaasOB @Sriraam_UTD @PacSymBiocomp

[Jacob Andreas @jacobandreas]

Guy van den Broeck: lots of people are building neural models with symbolic constrains---we just hide the symbolic parts