Publications

Publications

Selected papers with a short summary, the full list by year, and talks. For citation counts, see Google Scholar.

UNDER REVIEW

PCG-VAE: Learning Causal Representations under Partial Structural Knowledge

Emam Hossain, Md Osman Gani

Under review, CVPR 2027

Experts mark each causal link as known, forbidden, or unknown, and the model learns only the unknown ones. It recovers the hidden structure exactly in all 60 runs, compared with 3 of 20 for full discovery.

Paper coming soon
UNDER REVIEW

VLM-Gauge: Pricing Vision-Language Supervision for Causal Representation Learning

Emam Hossain, Md Osman Gani

Under review, ICLR 2027

Uses a vision-language model as the annotator and prices its labels before training: a label of quality J is worth about J² human labels. Labels on 17,000 images beat human labels on 4,000, and twelve attributes are supervised jointly.

Paper coming soon
ICMLA 2025

Causal Time Series Modeling of Supraglacial Lake Evolution in Greenland under Distribution Shift

Emam Hossain, Muhammad Hasan Ferdous, Devon Dunmire, Aneesh Subramanian, Md Osman Gani

IEEE ICMLA 2025

Lag-aware causal graphs choose the predictors. Causal models beat correlation-based baselines in all six leave-one-basin-out tests, with accuracy gains of up to 12.59%.

KDD 2025

TimeGraph: Synthetic Benchmark Datasets for Robust Time-Series Causal Discovery

Muhammad Hasan Ferdous, Emam Hossain, Md Osman Gani

ACM SIGKDD 2025, Datasets and Benchmarks Track

Benchmarks with known causal graphs plus real-world trends, seasonality, missing data, and hidden confounders. Under nonlinear dynamics, standard methods often find none of the true links.

TMLR 2023

A Survey on Causal Discovery Methods for I.I.D. and Time Series Data

Uzma Hasan, Emam Hossain, Md Osman Gani

Transactions on Machine Learning Research, 2023 (Survey Certification)

Maps what each family of causal discovery methods assumes, so a practitioner can see which assumptions their data break.

PolDS 2026

Supraglacial Lake Fate Is Knowable Long Before the Season Ends

Emam Hossain, Md Osman Gani

ACM SIGSPATIAL Workshop on Polar Data Science (PolDS) 2026

How early can we trust a prediction? Rapid lake drainage becomes predictable by 15 July, 92 days before a full-season model can be applied, and the ordering holds across learners and basins.

Under review and in preparation

2026

2025

2024

2023

2022

2021

2020

2019

† Undergraduate student I advised. Numbers count peer-reviewed papers from the first, in 2019.

Get in touch

I am always happy to talk about research, collaborations, and teaching.

Department of Information Systems, UMBC
Baltimore, Maryland