Skip to content
ICLR2023: Our paper on adapting pre-trained representations to ensure generalization and safety on downstream tasks accepted as a Spotlight. [ Paper][Code]
#WACV2023: Improving generalization of meta learners via constrastively-trained knowledge graph bridges — SoTA performance on few-shot dataset generalization [ Paper][ Code]
#WACV2023: Diversity or Adversity? What is more critical for domain generalization? Our new paper answers this question. hint: adversarially trained diverse augmentations is the trick [ Paper][ Code]
New Paper Alert: Solving severely ill-posed problems in CT imaging is valuable in a variety of applications including medical imaging and security. We introduce DOLCE, a conditional diffusion model, that produces state-of-the-art performance in limited angle CT reconstruction. [ Paper] Presenting two papers at
#NeurIPS2022: (i) Single-model uncertainty estimation using stochastic data centering ( Spotlight) and Analyzing Data-Centric Properties for Contrastive Learning on Graphs
#ACML2022: Interested in an effective OOD detector for your vision model? Try AMP that uses a neural network anchoring based uncertainty estimates for prediction calibration [ Paper][ Code]
#ACML2022: Fully test-time adaptation meets domain alignment! Check out CATTAn for adapting vision models at test-time under real-world distribution shifts [ Paper][ Code]
New Paper Alert: Zero-shot multi-domain generalization is challenging! We make an important finding that, the choice of domain grouping actually matters. Our new algorithm DReaME automatically discovers domain labels from multi-source data for optimal generalization. [ Paper][ Code] Received the
LLNL Director’s award for best publications during the year 2021 (Designing counterfactual generators from Neurips 2021 and Self-training for chest-xray classification from SPIE 2021) Nominated to attend the
#YoungLeadersProgram organized as part of the STS annual forum (Kyoto, Japan)! Presented a lecture at Microsoft Research on knowledge-aware deep learning [
Slides] Invited talk at Raytheon on OOD generalization and model safety [
Slides] Codes for our new OOD detection approach (AMP) released! Achieves state-of-the-art performance in vision benchmarks.
Github repo Look out for our papers at ICML2022 Workshops next week –
Principles of Distribution Shifts, Updatable ML, Interpretable ML in Healthcare, Healthare AI and Covid-19
New Paper Alert: In our new ICML2022 paper, we introduce SPHInX, a new GAN inversion technique, that effectively inverts OOD onto StyleGAN latent spaces. Presented two papers at ICASSP 2022 –
Attribute discovery in StyleGANs and Predicting generalization gap using anchoring. Check out our
journal article on transfer learning that was recently published in Machine Learning Science and Technology. #AI4Science New research in Interpretable ML for Healthcare — TraCE appears in Nature Scientific Reports. [
Paper] Interested in training machine-learning ready HPC ensembles in sciences? Check out our
new article in the Future Generation Computer Systems. Invited talk at AME digital culture series on “Grounding Deep Models via Uncertainty Characterization” [
Video][Slides]. Co-organized a mini-symposium with Sandeep Madireddy and Prasanna Balaprakash on “
Robust and Efficient Probabilistic Deep Learning for Scientific data and Beyond” at the SIAM UQ 2022 Conference. Three new papers accepted to the
Distribution Shifts workshop at Neurips 2021 – designing multi-domain ensembles, the role of domain-relabeling in generalization and unsupervised attribute alignment. Paper on building scientifically meaningful generative models for inertial confinement fusion to be presented at
ML4Physical Sciences workshop (Neurips 2021). Deep inversion-based Counterfactual reasoning with computer vision models accepted at Neurips 2021.
News feature on our recent work on building effective deep models with limited labeled data.
Best Paper Award at the SPIE 2021 Medical Imaging Conference for our paper on Self-Training with Improved Regularization for Sample-Efficient Chest X-Ray Classification!
Invited talk at the AIMI group in Stanford – Presented work on using prediction calibration to improve clinical diagnosis models.
News article on recent work on Learn-by-Calibrating for building high-fidelity scientific emulators. Our paper on calibration-driven learning featured on the
Nature Communications Editor’s Highlights!! New paper accepted for presentation at ICASSP 2021 – “Using deep image priors to generate counterfactual explanations” [
preprint] Three papers accepted to AAAI 2021 — robust explanations via loss estimation [
preprint], uncertainty matching Graph Neural Networks [ preprint] and attribute-guided adversarial augmentation [ preprint]. Presenting our work on connecting sample design and generalization in ML in Neurips 2020. [
paper] Work on using prediction calibration as a training objective for building regression models in scientific problems published in Nature Communications. [
paper] Need to build chest X-ray based diagnostic models with limited data? Check out our recently accepted SPIE paper!! [
, a general deep architecture for time-varying clinical data (ECG/EEG/EHR), published in Nature Scientific Reports. [ paper] [ code] Received the LLNL Director’s 2020 Early Career Recognition award.
Our paper on “A Statistical Mechanics Framework for Task-Agnostic Sample Design” accepted at Neurips 2020.
Invited Talk at the NNSA Next-Gen AI for Proliferation Detection Meeting on AI explainability [
slides] DOE Proposal on the integration of knowledge graphs into predictive modeling awarded.
Our paper on using prediction calibration to obtain reliable models in healthcare AI accepted for presentation at the
UNSURE workshop, MICCAI 2020. Paper on
Function-preserving linear projections (FPP) for high dimensional scientific data accepted for publication in the journal Machine Learning: Science and Technology. [ code] Our paper on
unsupervised audio source separation using GAN priors accepted for presentation at Interspeech 2020. [ code] Presented our paper on
Task-agnostic sample design in the Workshop on Real World Experiment Design and Active Learning at ICML 2020.
News article about our recent PNAS paper on using deep learning for surrogate modeling in scientific applications.
Feature article on AI-based analysis of clinical diagnosis models and COVID-19 infections. New podcast altert: DataSkeptic podcast where I talked about some recent work on interpretability in healthcare AI [
Apple] [ Spotify]. A
preprint of our work at LLNL on designing accurate emulators for scientific processes that is currently under consideration at Nature Communications. A
new approach to build reliable and interpretable deep models for Healthcare AI. In collaboration with IBM Research and Arizona State University. Paper on
building accurate neural network surrogates for inertial confinement fusion published in Proceedings of the National Academy of Sciences. [ code]
Coverage-based designs for hyper-parameter optimization in neural nets accepted for publication at IEEE Transactions on Neural Networks and Learning Systems. Our paper on
MimicGAN, an easy and effective way to robustly project onto the image manifold, is accepted to IJCV Special Issue on GANs.
Listen to our presentation on Learn-by-Calibrating at IEEE ICASSP 2020. New paper alert: How does prediction calibration affect the performance of lottery tickets? Our
new paper answers this question! A regularized GAT model for
robust semi-supervised learning under node injection attacks accepted as an oral talk at ICASSP 2020. Our paper
that explores the use of a prior-free calibration objective for training regression models accepted at ICASSP 2020. Learn-by-calibrating At AAAI presenting our work on building calibrated deep models for regression, time-series forecasting and object localization [
poster]. Presented our paper on weakly supervised instance labeling in histopathology images at ICMLA 2019 [
Slides]. Paper on uncertainty quantification with deep neural networks accepted at AAAI 2020.
technical report for my recently completed DOE-funded project on High-Dimensional Spectral Sampling. I will be presenting our paper on “
Improving Deep Embeddings for Inferencing with Multi-layered Graphs” in the Deep Graph Learning: Methodologies and Applications Workshop at IEEE Big Data 2019. Recieved the
DOE-ASCR Artificial Intelligence research grant to develop uncertainty quantification methods for deep learning. Four papers accepted for presentation at the Neurips 2019 Workshops, ML for Physical Sciences, Deep inverse and Graph Representation Learning.
New paper alert: Learning interpretable linear embeddings using
function preserving projections. Co-organized the 2nd
Applied Math Visioning workshop in New Mexico. Paper on
weakly supervised instance labeling in histopathological images accepted for oral presentation at ICMLA 2019.
Best paper award at the KDD Applied Data Science for Healthcare workshop. New paper alert: Do deep learning models in the clinical domain generalize under complex domain and task shifts?
Our work explores disease landscapes to characterize this behavior. New preprint on
unsupervised domain adaptation based on classical subspace analysis. Significant performance improvements over SOTA. State-of-the-art results obtained in audio source sepration via
multi-scale feature learning using dilated dense U-Nets. New results on
hyper-parameter optimization. Coverage-based sample designs identify highly-optimal configurations.
Updated version of the multi-layered graph attention models paper submitted. Co-organized the
1st DOE Applied Math Visioning Workshop and participated in exciting conversations about the future of ML/Data Science.
NVIDIA blog features results from our recent work on generative models.
Highlights from my research in the Computation Newsletter.