K
Kahini Wadhawan
Researcher at IBM
Publications - 15
Citations - 450
Kahini Wadhawan is an academic researcher from IBM. The author has contributed to research in topics: Domain (software engineering) & Dialog box. The author has an hindex of 7, co-authored 13 publications receiving 280 citations.
Papers
More filters
Proceedings Article
Co-regularized Alignment for Unsupervised Domain Adaptation
Abhishek Kumar,Prasanna Sattigeri,Kahini Wadhawan,Leonid Karlinsky,Rogerio Feris,Bill Freeman,Gregory W. Wornell +6 more
TL;DR: Co-regularized domain alignment as mentioned in this paper constructs multiple diverse feature spaces and aligns source and target distributions in each of them individually, while encouraging that alignments agree with each other with regard to the class predictions on the unlabeled target examples.
Journal ArticleDOI
Accelerated antimicrobial discovery via deep generative models and molecular dynamics simulations.
Payel Das,Payel Das,Tom Sercu,Tom Sercu,Kahini Wadhawan,Inkit Padhi,Sebastian Gehrmann,Sebastian Gehrmann,Flaviu Cipcigan,Vijil Chenthamarakshan,Hendrik Strobelt,Cicero Nogueira dos Santos,Cicero Nogueira dos Santos,Pin-Yu Chen,Yi Yan Yang,Jeremy P. K. Tan,James L. Hedrick,Jason Crain,Jason Crain,Aleksandra Mojsilovic +19 more
TL;DR: In this article, a computational method leveraging deep learning and molecular dynamics simulations enables the rapid discovery of antimicrobial peptides with low toxicity and with high potency against diverse Gram-positive and Gram-negative pathogens.
Posted Content
Co-regularized Alignment for Unsupervised Domain Adaptation
Abhishek Kumar,Prasanna Sattigeri,Kahini Wadhawan,Leonid Karlinsky,Rogerio Feris,William T. Freeman,Gregory W. Wornell +6 more
TL;DR: Co-regularized domain alignment as mentioned in this paper constructs multiple diverse feature spaces and aligns source and target distributions in each of them individually, while encouraging that alignments agree with each other with regard to the class predictions on the unlabeled target examples.
Posted Content
Optimizing Molecules using Efficient Queries from Property Evaluations.
TL;DR: QMO is proposed, a generic query-based molecule optimization framework that exploits latent embeddings from a molecule autoencoder that improves the desired properties of an input molecule based on efficient queries, guided by a set of molecular property predictions and evaluation metrics.
Posted Content
PepCVAE: Semi-Supervised Targeted Design of Antimicrobial Peptide Sequences.
Payel Das,Kahini Wadhawan,Oscar Chang,Tom Sercu,Cicero Nogueira dos Santos,Matthew Riemer,Inkit Padhi,Vijil Chenthamarakshan,Aleksandra Mojsilovic +8 more
TL;DR: A peptide generation framework PepCVAE, based on a semi-supervised variational autoencoder (VAE) model, for designing novel antimicrobial peptide (AMP) sequences that generates novel AMP sequences with higher long-range diversity, while being closer to the training distribution of biological peptides.