Krishnan Raghavan

My primary research agenda is to develop a mathematical characterization of machine learning~(ML) models, their learning/training behavior and the associated precision achieved by them. Towards this end, I study the two broad facets of ML: theory; through the eyes of tools from systems theory, statistics and optimization; and applied; by building AI/ML models to solve key problems in nuclear physics, material science, HPC and more recently climate.

Alt text

I love rock climbing and cycling. My favorite books include fantasy fiction and science.

Favorite fantasy fiction serries is Malazan Book of the Fallen by Steven Errikson

My favorite science book is What is Life? by Erwin Schrödinger.

selected publications

Continual Learning via Dynamic Programming

Krishnan Raghavan, and Prasanna Balaprakash

In International Conference on Pattern Recognition, 2022

Bib

@inproceedings{raghavanContinualLearningDynamic2022,
  title = {Continual Learning via Dynamic Programming},
  booktitle = {International {{Conference}} on {{Pattern Recognition}}},
  author = {Raghavan, Krishnan and Balaprakash, Prasanna},
  year = {2022},
  url = {https://ieeexplore.ieee.org/document/9956042},
}

Formalizing the Generalization-Forgetting Trade-off in Continual Learning

Krishnan Raghavan, and Prasanna Balaprakash

In Advances in Neural Information Processing Systems (AR:20), 2021

Bib

@inproceedings{raghavanFormalizingGeneralizationforgettingTradeoff2021,
  title = {Formalizing the Generalization-Forgetting Trade-off in Continual Learning},
  booktitle = {Advances in {{Neural Information Processing Systems}} ({{AR}}:20)},
  author = {Raghavan, Krishnan and Balaprakash, Prasanna},
  year = {2021},
  volume = {34},
  pages = {17284--17297},
}

Learning Continually on a Sequence of Graph – The Dynamical System Way

Krishnan Raghavan, and Prasanna Balaprakash

2023

Bib

@online{raghavanLearningContinuallySequence2023,
  title = {Learning {{Continually}} on a {{Sequence}} of {{Graph}} -- {{The Dynamical System Way}}},
  author = {Raghavan, Krishnan and Balaprakash, Prasanna},
  year = {2023},
  doi = {SIAM Journal on Mathematics of Data Science},
  pubstate = {preprint},
}

Automated Continual Learning of Defect Identification in Coherent Diffraction Imaging

Orcun Yildiz, Henry Chan, Krishnan Raghavan, and 5 more authors

In IEEE/ACM International Workshop on Artificial Intelligence and Machine Learning for Scientific Applications (AI4S), 2022

Abs

X-ray Bragg coherent diffraction imaging (BCDI) is widely used for materials characterization. However, obtaining X-ray diffraction data is difficult and computationally intensive. Here, we introduce a machine learning approach to identify crystalline line defects in samples from the raw coherent diffraction data. To automate this process, we compose a workflow coupling coherent diffraction data generation with training and inference of deep neural network defect classifiers. In particular, we adopt a continual learning approach, where we generate training and inference data as needed based on the accuracy of the defect classifier instead of all training data generated a priori. The results show that our approach improves the accuracy of defect classifiers while using much fewer samples of data.