Skip to content
AI News HubLIVE
More
Summary only

Interview with Deepika Vemuri: interpretability and concept-based learning

Deepika Vemuri, a PhD student at IIT Hyderabad, discusses her research on interpretability and concept-based learning. She explores using logic for concept-to-class mapping and aligning neural network depth with semantic hierarchy via formal concept analysis. She also shares her internship at Vector Institute and future plans.

SourceAIhubAuthor: Lucy Smith
Interview with Deepika Vemuri: interpretability and concept-based learning
Report an error

The correction channel is not available yet. You can copy the article reference below for later.

Correction instructions

Summary only

Under the source content permission, this page provides an editorial summary, key points, and a link to the original article.

Key points and analysis

Article intelligence

EngineersAdvanced

Key points

  • Vemuri uses fuzzy logic gates to model concept combinations, improving accuracy and interpretability.
  • She leverages formal concept analysis to build concept lattices, guiding networks to learn concepts at different abstraction levels across layers.
  • She proposes a worst-case metric for concept models measuring confidence improvement after removing misleading concepts.
  • Currently interning at Vector Institute, she works on representing video as concept graphs for improved VideoQA.

Highlights and analysis are generated automatically and may contain errors. Check the original source.

More recent stories on this topic · Research