PhD in CS at Oregon State University advised by Xiao Fu. Research in unsupervised domain translation, generative models, and multiview/multimodal learning.
Applied Scientist Intern at Amazon. Graph semi-supervised learning for account takeover detection.
Computer Vision Research Intern at Samsung Research America. Research in personalization of text to image generative models with Gopal Sharma, Luowei Zhou, and Suren Kumar.
MS in CS at OSU (advisor: Xiao Fu). Research in ML/DL/RL for sensing, communication and federated learning, funded by Intel and NSF (MLWiNS).
Co‑founded Paaila Technology in Nepal. Led development of waiter robots.
BE in ECE at IOE, Pulchowk Campus, Tribhuvan University, Nepal.
Team Member of Robotics Club. Participated in ABU Robocon 2015 & 2016. Our team won multiple awards.
International Conference on Machine Learning (ICML) 2025
Diversified distribution matching (DDM) finds a unified translation function mapping a diverse collection of conditional source distributions to their target counterparts. DDM was proposed to resolve content misalignment issues in unpaired domain translation, achieving translation identifiability. However, DDM has only been implemented using GANs due to its constraints on the translation function. This work introduces diversified flow matching (DFM), an ODE-based framework for DDM.
International Conference on Learning Representations (ICLR) 2025
This work develops a novel framework for identifying latent content and style variables from unaligned multi-domain data, a crucial challenge in domain translation and generative modeling. We introduce cross-domain latent distribution matching (LDM) along with sparsity constraint to identify the latent content and style variables without latent dimension information.
Neural Information Processing Systems (NeurIPS) 2024
This work investigates shared component identifiability from multi-modal linear mixtures with unaligned cross-modality samples, extending beyond previous research on aligned samples. We propose a distribution divergence minimization-based loss and derive sufficient conditions for shared component identifiability.
International Conference on Learning Representations (ICLR) 2024
Unsupervised domain translation is the problem of learning a mapping between two domains without paired data. Existing methods generally tackle the problem with distribution matching objectives. A fundamental issue with these approaches is the "non-identifiability" of translation maps due to the existence of multiple solutions. Our work proposes a simple yet provable solution to address this issue.
IEEE SAM Workshop 2024
Optical Coherence Tomography (OCT) provides detailed cross-sectional images of coronary arteries, but cost-effective systems produce only low-resolution images. Our work proposes a translation identifiability-guided framework using a diversified distribution matching module for OCT super-resolution.
IEEE Transactions on Signal Processing (TSP) 2023
Joint Beamforming and Antenna selection is a mix of non-convex continuous and combinatorial optimization problem. We address this issue by first proposing an optimal Branch and Bound (B&B) algorithm for the problem. To address the potential scalability issue of the B&B, we propose a GNN based policy learnt via imitation learning.
IEEE Transactions on Signal Processing (TSP) 2023
Classic and deep learning-based GCCA algorithms seek low-dimensional common representations of data entities from multiple "views" (e.g., audio and image) using linear transformations and neural networks, respectively. This work puts forth a convergence-guaranteed communication-efficient federated learning framework for both linear and deep GCCA under the Maximum Variance (MAX-VAR) formulation.
IEEE Transactions on Signal Processing (TSP) 2023
Spectrum cartography (SC), also known as radio map estimation (RME), aims at crafting multi-domain (e.g., frequency and space) radio power propagation maps from limited sensor measurements. This work puts forth a quantized SC framework that generalizes the BTD and DGM-based SC to scenarios where heavily quantized sensor measurements are used.
IEEE Transactions on Signal Processing (TSP) 2022
The spectrum cartography (SC) technique constructs multi-domain (e.g., frequency, space, and time) radio frequency (RF) maps from limited measurements, which can be viewed as an ill-posed tensor completion problem. In this work, an emitter radio map disaggregation-based approach is proposed, under which only individual emitters radio maps are modeled by DNNs.
A note on how to view flow matching and diffusion under the same framework, resulting in great flexibility in designing probability path, training score or vector field, and sampling using SDE or ODE.
A short note on understanding the diffusion and flow matching objectives and their solutions.
A brief survey into joint representation learning of text and images.
A distilled explanation of the flow matching / stochastic interpolant /rectified flow. It features an easy-to-follow proof (not found in original paper) of the flow-matching objective with stochastic interpolants.
A beginner's introduction to ICA.
