Research

<Under construction>

What are the computations underlying social interaction recognition in visual scenes?

Model paper: Relational visual representations underlie human social interaction recognition.

Previous Research

• Understanding multi-sensory perception using graph theory, EEG signals and the McGurk effect
• Deep reinforcement learning model for dynamic optimization of budget-constrained campaign allocation
• Network Analysis of Food-Disease Associations

Publications & Select Presentations

Publications

- Malik, M., Isik, L., Relational visual representations underlie human social interaction recognition., Nature Communications 14, 7317 (2023).

- M. Malik, G. Gupta, L. Vig, and G. Shroff, BCQ4DCA: Budget Constrained Deep Q-Network for Dynamic Campaign Allocation in Computational Advertising, IEEE International Joint Conference on Neural Networks, 2021 (IJCNN '21).

- Yashaswi Rauthan, Vatsala Singh, Rishabh Agrawal, Satej Kadlay, Niranjan Pedanekar, Shirish Karande, Manasi Malik, and Iaphi Tariang, Avoid Crowding in the Battlefield: Semantic Placement of Social Messages in Entertainment Programs, International Workshop on AI for Smart TV Content Production, Access and Delivery (AI4TV '20).

- Manasi Malik, Ganesh Bagler, and Arpan Banerjee. Network analysis of neuro-cognitive processes: studying McGurk effect using EEG data, IIITD, 2019.

Talks

- Symposium, Social & Affective Neuroscience Society (SANS) (2024)

- Talk sessions, Vision Sciences Society (VSS), Florida, USA (2022)

- Lab Meeting, Social Computation Representation And Prediction Lab (SCRAP), Dartmouth (2024)

- Lab Meeting, JHU Social & Cognitive Origins Group (2023)

- Lab Meeting, MIT Computational Cognitive Science group (2022)

- Brown Bag Talk, JHU Cognitive Science (2022)

Patents

- Gupta, G., Vig, L., Shroff, G., & Malik, M. (2024). Budget constrained deep Q-network for dynamic campaign allocation in computational advertising. U.S. Patent No. 11,915,262.

Posters

- Manasi Malik, Minjae Kim, Shari Liu, Tianmin Shu, Leyla Isik, Investigating the neural computations underlying visual social inference with graph neural networks, Conference on Cognitive Computational Neuroscience (CCN'24), Boston, USA.

- Manasi Malik, Leyla Isik, Human Social Interaction Judgements are Uniquely Explained by both Bottom-up Graph Neural Networks and Generative Inverse Planning Models, Conference on Cognitive Computational Neuroscience (CCN'23), Oxford, UK.

- Manasi Malik, Leyla Isik, Human Social Interaction Judgements are Uniquely Explained by both Bottom-up Graph Neural Networks and Generative Inverse Planning Models, Johns Hopkins AI-X Foundry Fall 2023 Symposium, Baltimore, USA.

- Manasi Malik, Leyla Isik, Both Purely Visual and Simulation-based Models Uniquely Explain Human Social Interaction Judgements, Vision Sciences Society (VSS '23), Florida, USA.

- Manasi Malik, Leyla Isik, Social Inference from Relational Visual Information: An Investigation with Graph Neural Network Models, Conference on Cognitive Computational Neuroscience (CCN'22), San Francisco, USA (poster).