Causal Machine Learning. Neuro-Symbolic AI. Healthcare. Human in the Loop Learning. Probabilistic Deep Models.
I am an Assistant Professor at TU Eindhoven in the Uncertainity in Artificial Intelligence group.
Previously, I was a DEPTH research group leader on Causality And neUro-Symbolic artificial intElligence (CAUSE), under The Hessian Center for Artificial Intelligence (hessian.AI) and TU Darmstadt.
I was also a Postdoctoral Researcher under Prof. Dr. Kristian Kersting at TU Darmstadt in the Artificial Intelligence and Machine Learning Lab.
Ph.D. in Computer Science from University of Texas, Dallas. My PhD thesis focused on learning effective models from noisy, heterogeneous and multi-relational healthcare data.
MS in Computer Science from Indiana University, Bloomington. (My master's thesis focused on creating novel features from galaxy images for spiral shape detection.)
Email: devendra.dhami [at] cs.tu-darmstadt [dot] de ; Phone: +49 1523 7919 263
Research interests: Causal Machine Learning ● Neuro-Symbolic AI ● ML in healthcare ● Probabilistic Deep Models ● Human-in-the-loop Deep learning ● Graphical Models ● Knowledge Graphs ● Scaling in Probabilistic Logic models.
My Bachelors was completed from Sir MVIT Bangalore (@Visvesvaraya Technological University) with a gold medal and an academic excellence award. Prior to pursuing graduate studies, I have worked with Hewlett-Packard. I also write short stories as a hobby and am also an avid reader.
Professional Services
1. Electronics Publishing Editor: Journal of Artificial Intelligence Research (JAIR)
2. Action Editor: Transactions on Machine Learning Research (TMLR)
3. Proceedings Chair: SDM 2020
4. Editorial Board: Frontiers in Artificial Intelligence
5. Area Chair: Neuro-Symbolic AI Workshop, NeurIPS 2024
6. PC member: Pacific Symposium on Biocomputing 2019,2020,2021; IJCAI 2020, 2022; SDM 2020,2021; AAAI 2021 (top 25%),2022; ICML 2021,2022; UAI 2021,2022; NeurIPS 2021,2022,2023 (Top reviewer); ACM Multimedia 2021, 2022; ICMLA 2021; ICLR 2022; AISTATS 2022; IJCAI (Demo Track) 2022, IJCAI (Survey Track) 2022, CoDS-COMADS 2020
7. Reviewer (Journal): Journal of Machine Learning Research (JMLR), Big Data Journal, Data Mining and Knowledge Discovery (DAMI), Journal of Artificial Intelligence Research (JAIR), Machine Learning Journal (MLJ), IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), Frontiers in Artificial Intelligence, iScience: Cell Press; NeuroImage
8. Volunteer: ICDE 2020
9. REU Mentor: Mentored 2 undergraduate researchers for Indiana University's Proactive Health Informatics "Research experiences for Undergraduates (REU) program"
News/Updates:
2 papers accepted in NeurIPS 2024: "DeiSAM: Segment Anything with Deictic Prompting" and "Graph Neural Networks Need Cluster-Normalize-Activate Modules".
I received the hessian.AI DEPTH Alumnus award. This helps me to travel to Darmstadt to keep advising my PhD students.
Our paper 'Machine Learning meets Kepler: Inverting Kepler’s Equation for All vs All Conjunction Analysis' is accepted to Machine Learning: Science and Technology (MLST).
Our paper 'Neuro-Symbolic Predicate Invention: Learning Relational Concepts from Visual Scenes' is accepted to the Neurosymbolic Artificial Intelligence journal.
2 papers accepted to UAI.
Our paper 'Towards Probabilistic Clearance, Explanation and Optimization' is accepted to the International Conference on Unmanned Aircraft Systems (ICUAS).
Our paper 'Structural Causal Models Reveal Confounder Bias in Linear Program Modelling' is accepted to the Asian Conference on Machine Learning (ACML) journal track.
2 papers accepted to NeurIPS.
Our paper 'Scalable Neural-Probabilistic Answer Set Programming' will appear in Journal for Artificial Intelliegnce Research (JAIR).
Our paper 'Causal Parrots: Large Language Models May Talk Causality But Are Not Causal' will appear in Transactions on Machine Learning Research (TMLR).
Our paper 'Vision Relation Transformer for Unbiased Scene Graph Generation' will appear in International Conference on Computer Vision (ICCV).
I am a co-organiser of AAAI 2023 bridge on 'Continual Causality' which aims to bring researchers from continual learning and causality together.
Our paper 'alphaILP: Thinking Visual Scenes as Differentiable Logic Programs' will appear in Machine Learning Journal and IJCLR 2022.
Our paper 'Predictive Whittle Networks for Time Series' is accepted to UAI 2022.
2 papers on probabilistic logic languages accepted to International Conference on Principles of Knowledge Representation and Reasoning (KR).
Our paper on training CTGANs in a differentially private way is accepted to Artificial Intelligence in Medicine (AIME) 2022.
Our team '51 Billion Shades of Carbon' won the Climathon, Darmstadt. It is an yearly hackathon that seeks submissions to combat climate change.
Our paper 'Interventional Sum-Product Networks: Causal Inference with Tractable Probabilistic Models' is accepted to NeurIPS 2021.
Our paper 'A Statistical Relational Approach to Learning Distance-based GCNs' is accepted to Tenth International Workshop on Statistical Relational AI.
2 papers accepted to the International Conference on Inductive Logic Programming, ILP2020-21 @ 1st International Joint Conference on Learning & Reasoning (IJCLR).
Our paper 'Interventional Sum-Product Networks: Causal Inference with Tractable Probabilistic Models' is accepted to the Tractable Probabilistic Modeling (TPM) workshop @UAI.
Our paper 'Beyond Simple Images: Human Knowledge-Guided GANs for Clinical Data Generation' is accepted to International Conference on Principles of Knowledge Representation and Reasoning (KR).
I gave a talk on 'Distance based Graph Convolutional Networks' at the Cyberinfrastructure for Network Science Center @ Indiana University Bloomington research symposium.
Our paper 'Predicting Drug-Drug Interactions from Heterogeneous Data: An Embedding Approach' is accepted to Artificial Intelligence in Medicine (AIME).
I am the new Electronics Publishing Editor for Journal of Artificial Intelligence Research (JAIR).
Our paper 'The Curious Case of Stacking Boosted Relational Dependency Networks' is accepted to I Can't Believe It's Not Better (ICBINB) workshop @NeurIPS as a spotlight presentation.
Our paper 'Human-Guided Learning of Column Networks: Knowledge Injection for Relational Deep Learning' is accepted to ACM India CODS-COMAD 2021.
Our paper 'Knowledge Intensive Learning of Generative Adversarial Networks' received the best paper award at KiML workshop @KDD.
The paper 'Knowledge Intensive Learning of Generative Adversarial Networks' is accepted to the KiML workshop @KDD.
I am in the editorial board for 'Frontiers in Artificial Intelligence'.
I have graduated with a PhD and will be joining TU Darmstdat as a Postdoc in September 2020.