Deepak Venugopal

I am an associate professor and the graduate coordinator in the department of Computer Science at University of Memphis. I obtained my PhD in computer science from UT-Dallas.

Broadly, my research interest is in the fields of Artificial Intelligence and Machine Learning. Currently, I am interested in Neuro-Symbolic AI (integrating neural networks with symbolic AI) and applications particularly in educational technologies.

News

• New paper on representation learning to understand Math strategies to be presented at LAK 2025

• New paper on disentangling fine-tuned examples from pre-training in captioning models presented at IEEE BigData workshop on Large Language and Foundation Models 2024

• New paper on learning Neuro-Symbolic models with Hybrid Markov Logic and applications in cognitive models for education to be presented at a AAAI-bridge (congnitive science and Neuro-Symbolic AI) 2025

• New paper at ACM Learning @ Scale 2024 on representing math strategies using BERT

• Co-authored book on Dimensionality Reduction available at the following link https://link.springer.com/book/10.1007/978-3-031-05371-9

Publications

1. Abisha Thapa Magar, Stephen E. Fancsali, Vasile Rus, April Murphy, Steve Ritter, Deepak Venugopal. "Can A Language Model Represent Math Strategies?": Learning Math Strategies from Big Data using BERT, International Learning Analytics and Knowledge Conference (LAK), 2025.

2. Monika Shah, Somdeb Sarkhel, Deepak Venugopal, Disentangling Fine-Tuning from Pre-Training in Visual Captioning with Hybrid Markov Logic, IEEE Big Data workshop on Large Language and Foundation Models, 2024.

3. Abisha Thapa Magar, Stephen E. Fancsali, Vasile Rus, April Murphy, Steve Ritter, Deepak Venugopal. Learning Representations for Math Strategies using BERT, ACM Learning @ Scale 2024.

4. Abisha Thapa Magar, Anup Shakya, Somdeb Sarkhel and Deepak Venugopal, Verifying Relational Explanations: A Probabilistic Approach, IEEE Big Data 2023

5. Anup Shakya, Abisha Thapa Magar, Somdeb Sarkhel and Deepak Venugopal, On the Verification of Embeddings with Hybrid Markov Logic, ICDM 2023

6. Anup Shakya, Vasile Rus and Deepak Venugopal, Scalable and Equitable Math Problem Solving Strategy Prediction in Big Educational Data, EDM 2023 [pdf] [code]

7. Monika Shah, Somdeb Sarkhel, Deepak Venugopal, Evaluating Captioning Models using Markov Logic Networks, IEEE Big Data 2022.

8. William Britton, Somdeb Sarkhel, Deepak Venugopal, Question Modifiers in Visual Question Answering, LREC 2022. [pdf]

9. Mohammad Maminur Islam, Somdeb Sarkhel and Deepak Venugopal, Contrastive Learning in Neural Tensor Networks using Asymmetric Examples, IEEE Big Data, 2021. [code] (https://ieeexplore.ieee.org/document/9671631)

10. Khan Md. Al Farabi, Somdeb Sarkhel, Sanorita Dey and Deepak Venugopal, Interpretable Explanations for Probabilistic Inference in Markov Logic, IEEE Big Data, 2021. (https://ieeexplore.ieee.org/document/9671572)

11. Anup Shakya, Vasile Rus and Deepak Venugopal, Student Strategy Prediction using a Neuro-Symbolic Approach, EDM 2021. [pdf] [code]

12. Mohammad Maminur Islam, Somdeb Sarkhel and Deepak Venugopal, Augmenting Deep Learning with Relational Knowledge from Markov Logic Networks, IEEE Big Data, 2020. https://ieeexplore.ieee.org/document/9378055. [code]

13. Anik Khan, Kishor Datta Gupta, Deepak Venugopal and Nirman Kumar (2020). CIDMP: Completely Interpretable Detection of Malaria Parasite in Red Blood Cells using Lower-dimensional Feature Space. International Joint Conference on Neural Networks (IJCNN). (https://ieeexplore.ieee.org/document/9206885)

14. Naveen Kumar, Deepak Venugopal, Liangfei Qiu and Subodha Kumar, Detecting Anomalous Online Reviewers: An Unsupervised Approach Using Mixture Models, Journal of Management Information Systems, Vol. 36 No. 4 2019. (https://www.tandfonline.com/doi/abs/10.1080/07421222.2019.1661089?journalCode=mmis20)

15. Khan Md. Al Farabi, Somdeb Sarkhel, Sanorita Dey and Deepak Venugopal, Fine-grained Explanations using Markov Logic, ECML/PKDD 2019 [pdf]

16. Craig Kelly, Somdeb Sarkhel and Deepak Venugopal, Adaptive Rao-Blackwellisation in Gibbs Sampling for Probabilistic Graphical Models, AISTATS, 2019 [pdf] [code]

17. Mohammad Maminur Islam, Somdeb Sarkhel and Deepak Venugopal, On Lifted Inference using Neural Embeddings, AAAI 2019 [pdf] [code]

18. Das, Saikat, A. Mahfouz, D. Venugopal, and S. Shiva. DDoS Intrusion Detection through Machine Learning Ensemble, IEEE International Conference on Software Quality, Reliability and Security Companion (QRS). (https://ieeexplore.ieee.org/document/8859416)

19. Mohammad Maminur Islam, Khan Mohammad Al Farabi, Somdeb Sarkhel and Deepak Venugopal, Scaling up Inference in MLNs with Spark, IEEE BigData 2018. (https://ieeexplore.ieee.org/document/8622607)

20. Naveen Kumar, Deepak Venugopal, Liangfei Qiu and Subodha Kumar, Detecting Review Manipulation on Online Platforms with Hierarchical Supervised Learning, 35 (1), Journal of Management Information Systems (JMIS). [https://www.tandfonline.com/doi/abs/10.1080/07421222.2018.1440758]

21. Khan Al Farabi, Somdeb Sarkhel and Deepak Venugopal, "Efficient Weight Learning in High-Dimensional Untied MLNs", AISTATS 2018 [pdf].

22. Mohammad Maminur Islam, Somdeb Sarkhel and Deepak Venugopal, "Learning Mixtures of MLNs", AAAI 2018 [pdf].

23. Deepak Venugopal, Advances in Inference Methods for Markov Logic Networks. IEEE Intelligent Informatics Bulletin 18(2): 13-19 (2017) [pdf]

24. Somdeb Sarkhel, Deepak Venugopal, Nicholas Ruozzi and Vibhav Gogate, "Efficient Inference for Untied MLNs", IJCAI 2017 [pdf]

25. Mohammad Maminur Islam, Mohammad Khan Al Farabi and Deepak Venugopal, "Adaptive Blocked Gibbs Sampling for Inference in Probabilistic Graphical Models", IJCNN, 2017. (https://ieeexplore.ieee.org/document/7965864)

26. Deepak Venugopal and Vasile Rus, "Joint Inference for Mode Identification in Tutorial Dialogues", COLING 2016 [pdf]

27. Jing Lu, Deepak Venugopal, Vibhav Gogate and Vincent Ng, "Joint Inference for Event Coreference Resolution", COLING 2016 [pdf]

28.  Deepak Venugopal, Somdeb Sarkhel and Kyle Cherry, “Non-parametric Domain Approximation for Scalable Gibbs Sampling in MLNs,” UAI 2016. [pdf] 

29. Somdeb Sarkhel, Deepak Venugopal, Tuan Anh Pham, Parag Singla and Vibhav Gogate, “Scalable Training of Markov Logic Networks using Approximate Counting,” AAAI 2016. [pdf] [extended-version]

30. Deepak Venugopal, Somdeb Sarkhel and Vibhav Gogate, Just Count the Satisfied Groundings: Scalable Local-Search and Sampling Based Inference in MLNs, In 29th AAAI Conference on Artificial Intelligence (AAAI), 2015. [pdf]

31. Deepak Venugopal and Vibhav Gogate, Scaling-up Importance Sampling for Markov Logic Networks, In 28th Annual Conference on Neural Information Processing Systems (NIPS), 2014. [pdf]

32. Somdeb Sarkhel, Deepak Venugopal, Parag Singla and Vibhav Gogate, An Integer Polynomial Programming Based Framework for Lifted MAP Inference, In 28th Annual Conference on Neural Information Processing Systems (NIPS), 2014. [pdf]

33. Deepak Venugopal, Chen Chen, Vibhav Gogate and Vincent Ng, Relieving the Computational Bottleneck: Joint Inference for Event Extraction with High-Dimensional Features, In Empirical Methods in Natural Language Processing Conference (EMNLP), 2014. [pdf]

34. Deepak Venugopal and Vibhav Gogate, Evidence-Based Clustering for Scalable Inference in Markov Logic, In European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML/PKDD), 2014. [pdf]

35. Somdeb Sarkhel, Deepak Venugopal, Parag Singla and Vibhav Gogate, Lifted MAP Inference for Markov Logic Networks, In 17th International Conference on Artificial Intelligence and Statistics (AISTATS), 2014. [pdf]

36. Deepak Venugopal and Vibhav Gogate, Dynamic Blocking and Collapsing for Gibbs Sampling, In 29th Conference on Uncertainty in Artificial Intelligence (UAI), 2013. [pdf]

37. Deepak Venugopal and Vibhav Gogate, GiSS: Combining Gibbs Sampling and SampleSearch for Inference in Mixed Probabilistic and Deterministic Graphical Models, In 27th AAAI Conference on Artificial Intelligence (AAAI), 2013. [pdf]

38. Deepak Venugopal and Vibhav Gogate, On Lifting the Gibbs Sampling Algorithm, In 26th Annual Conference on Neural Information Processing Systems (NIPS), 2012. [pdf]

39. Vibhav Gogate, Abhay Jha and Deepak Venugopal, Advances in Lifted Importance Sampling, In 26th AAAI Conference on Artificial Intelligence (AAAI), 2012. [pdf]

Peer-Reviewed Workshop Papers

1. Anup Shakya, Vasile Rus, Deepak Venugopal, A Neuro-Symbolic Approach for Student Strategy Prediction, RL4ED workshop@AAAI 2022.

2. Mohammad Maminur Islam, Somdeb Sarkhel and Deepak Venugopal, "Learning Embeddings for Approximate Lifted Inference in MLNs", NeurIPS Relational Representation learning (R2L) workshop, 2018

3. Christopher Kent and Deepak Venugopal, Fine-Grained Crime Prediction in an Urban Neighborhood, IEEE Conference on Smart Cities 2018 [poster] (https://ieeexplore.ieee.org/document/8656734)

4. Hari Charan Cheekati, Swaroop Goli and Deepak Venugopal, "MSpark: A Scalable Lifted Inference Pipeline for MLNs", In IJCAI-16 workshop on Statistical Relational Artificial Intelligence, 2016. 

5. Deepak Venugopal, Scaling-up Inference in Markov Logic, Extended Abstract, In AAAI-15 Doctoral Consortium, 2015. [pdf]

6. Deepak Venugopal and Vibhav Gogate, Evidence-Based Clustering for Scalable Inference in Markov Logic, In AAAI-14 Workshop on Statistical Relational Artificial Intelligence, 2014.

7. Deepak Venugopal and Vibhav Gogate, On Lifting the Gibbs Sampling Algorithm, In UAI-12 Workshop on Statistical Relational Artificial Intelligence, 2012.

Current Students

Anup Shakya (Ph.D.)

Monika Shah (Ph.D.)

Abisha Thapa (Ph.D.)

Alumni

Md. Maminur Islam, Ph.D.  (Trinity College)

Dissertation: Advances in Improving Scalability and Accuracy of MLNs using Symmetries

Khan Md. Al Farabi, Ph.D. (Augusta University)

Dissertation: Explanation Techniques using Markov Logic Networks

Prachi Jadhav, MS (Stacklox)

Thesis: Resampling Generative Models: An Empirical Study

William Britton, MS

Thesis: Question Modifiers In VQA: Evaluating Model Sensitivity

Craig Kelly, MS (Fedex)

Thesis: Parallel, Adaptive, Collapsed Gibbs Sampling

Tutorials

• AAAI 2018 [pdf]

• IJCAI 2017 [pdf]

• EMNLP 2016 [pdf]

RECENT Presentations

• Using AI to Understand the Impact of Math Strategy Use, AIMS Meeting (joint talk with Steve Ritter, Carnegie Learning), 2024

• Learning to Represent Math Strategies, UM Cognitive Science Seminar, 2024

• Verification of Embeddings using Hybrid Markov Logic, UM Cognitive Science Seminar, 2024

• Explanations using Markov Logic, Deepak Venugopal, UM Cognitive Science Seminar, 2022 [pdf]

• Student Strategy Prediction using Neuro-Symbolic AI, EDM 2021 [pdf] [video]

• Neuro-Symbolic AI: A scalable, Explainable Framework for Strategy Discovery from Big Edu-Data, LDI Workshop@EDM 2021 [pdf]

Teaching

• Introduction to Machine Learning

• Artificial intelligence

• Fundamentals of Data Science

• Database Systems

Service

Reviewer/PC: AISTATS, NIPS, ICML, AAAI, IJCAI, NAACL,ICLR,ECML/PKDD,TDKE, NNLS, AIJ, JAIR

Software

• 1. SATTN: Attention-based model for predicting student strategies equitably over different levels of mastery. Please consider citing (Shakya et al., EDM 2023) for this work.

  2. SSPM: Neuro-Symbolic model for predicting student strategies. Please consider citing (Shakya et al., EDM 2021) for this work.

  3. MNTN: Training NTNs using symmetries in the MLN. Please consider citing (Islam et al. IEEE Bigdata 2021) for this work.

  4. MLCNN: Combining MLNs with CNNs. Please consider citing (Islam et al., IEEE BigData 2020) for this work.

  5. Obj2Vec is an embedding for MLNs based on symmetries. Please consider citing (Islam et al., AAAI 2019) for this work.

  6. Grample  performs Parallel, Rao-Blackwellised Gibbs sampling for discrete PGMs. and is implemented in GO. Please consider citing Kelly et al. AISTATS 2019 paper if you use this software.

  7. Magician controls inference complexity and scales-up to large domains by combining lifted inference with approximations and advanced solution counters. The current software is a Beta-version and implements Gibbs sampling along with learning using Contrastive Divergence. Please consider citing the papers: Venugopal et al. AAAI 2015 and Sarkhel et al. AAAI 2016, if you use this software for your research.

  8. Alchemy-2 extends Alchemy (suite of inference/learning algorithms for Markov Logic) by implementing a number of lifted inference algorithms. Lifted inference algorithms differ from traditional propositional inference algorithms by performing inference at the first-order level as far as possible and propositionalizing only as needed. Lifted Inference algorithms therefore offer far greater scalability when compared to propositional algorithms.

Selected Co-Invented Patents (Granted)

Wireless intrusion prevention system and method