Publications

2024

1. NeurIPS

   DiGRAF: Diffeomorphic Graph-Adaptive Activation Function

   Mantri, Krishna Sri Ipsit, Wang, Xinzhi, Schönlieb, Carola-Bibiane, Ribeiro, Bruno, Bevilacqua, Beatrice, and Eliasof, Moshe

   In The Thirty-eighth Annual Conference on Neural Information Processing Systems 2024

   HTML PDF
2. NeurIPS, UniReps

   Rethinking Fine-tuning Through Geometric Perspective

   Mantri, Krishna Sri Ipsit, Eliasof, Moshe, Schönlieb, Carola-Bibiane, and Ribeiro, Bruno

   In UniReps: 2nd Edition of the Workshop on Unifying Representations in Neural Models at NeurIPS 2024

   HTML PDF

2023

1. WSDM

   Learning and Maximizing Influence in Social Networks Under Capacity Constraints

   Chakraborty, Pritish, Ranu, Sayan, Mantri, Krishna Sri Ipsit, and De, Abir

   In Proceedings of the Sixteenth ACM International Conference on Web Search and Data Mining 2023

   Abs Bib HTML PDF

   Influence maximization (IM) refers to the problem of finding a subset of nodes in a network through which we could maximize our reach to other nodes in the network. This set is often called the "seed set", and its constituent nodes maximize the social diffusion process. IM has previously been studied in various settings, including under a time deadline, subject to constraints such as that of budget or coverage, and even subject to measures other than the centrality of nodes. The solution approach has generally been to prove that the objective function is submodular, or has a submodular proxy, and thus has a close greedy approximation. In this paper, we explore a variant of the IM problem where we wish to reach out to and maximize the probability of infection of a small subset of bounded capacity K. We show that this problem does not exhibit the same submodular guarantees as the original IM problem, for which we resort to the theory of gamma-weakly submodular functions. Subsequently, we develop a greedy algorithm that maximizes our objective despite the lack of submodularity. We also develop a suitable learning model that out-competes baselines on the task of predicting the top-K infected nodes, given a seed set as input.

   @inproceedings{10.1145/3539597.3570433,
     abbr = {WSDM},
     bibtex_show = {true},
     author = {Chakraborty, Pritish and Ranu, Sayan and Mantri, Krishna Sri Ipsit and De, Abir},
     title = {Learning and Maximizing Influence in Social Networks Under Capacity Constraints},
     year = {2023},
     isbn = {9781450394079},
     publisher = {Association for Computing Machinery},
     address = {New York, NY, USA},
     url = {https://doi.org/10.1145/3539597.3570433},
     html = {https://doi.org/10.1145/3539597.3570433},
     pdf = {assets/pdf/wsdmfp0423-chakraborty1.pdf},
     doi = {10.1145/3539597.3570433},
     booktitle = {Proceedings of the Sixteenth ACM International Conference on Web Search and Data Mining},
     pages = {733–741},
     numpages = {9},
     keywords = {gamma-weakly submodular functions, deep submodular functions, influence maximization, graph neural networks},
     location = {Singapore, Singapore},
     series = {WSDM '23}
   }

2. ICLR

   STAGCN: Spatial-Temporal Attention Based Graph Convolutional Networks for COVID-19 Forecasting

   Sasikumar, Nevasini, and Mantri, Krishna Sri Ipsit

   In 2023 ICLR First Workshop on “Machine Learning & Global Health” 2023

   Bib HTML PDF

   @inproceedings{sasikumar2023stagcn,
     title = {{STAGCN}: Spatial-Temporal Attention Based Graph Convolutional Networks for {COVID}-19 Forecasting},
     author = {Sasikumar, Nevasini and Mantri, Krishna Sri Ipsit},
     booktitle = {2023 ICLR First Workshop on {\textquotedblleft}Machine Learning {\&} Global Health{\textquotedblright}},
     year = {2023},
     url = {https://openreview.net/forum?id=k0E_VMXLXl},
     html = {https://openreview.net/forum?id=k0E_VMXLXl},
     pdf = {https://openreview.net/pdf?id=k0E_VMXLXl},
     abbr = {ICLR},
     bibtex_show = {true}
   }

3. ICLR

   Attention Based Variational Graph Auto-Encoder (AVGAE)

   Sasikumar, Nevasini, and Mantri, Krishna Sri Ipsit

   In ICLR 2023, Tiny Papers 2023

   Bib HTML PDF

   @misc{sasikumar2023attention,
     title = {Attention Based Variational Graph Auto-Encoder ({AVGAE})},
     author = {Sasikumar, Nevasini and Mantri, Krishna Sri Ipsit},
     booktitle = {ICLR 2023, Tiny Papers},
     year = {2023},
     url = {https://openreview.net/forum?id=j1gj0ndrk1},
     html = {https://openreview.net/forum?id=j1gj0ndrk1},
     pdf = {https://openreview.net/pdf?id=j1gj0ndrk1},
     abbr = {ICLR},
     bibtex_show = {true}
   }

4. MIDL

   Synthetic Medical Image Generation Using Latent Diffusion Models and Large Language Models

   Mantri, Krishna Sri Ipsit, and Sasikumar, Nevasini

   In Medical Imaging with Deep Learning, short paper track 2023

   Bib HTML PDF

   @inproceedings{mantri2023synthetic,
     title = {Synthetic Medical Image Generation Using Latent Diffusion Models and Large Language Models},
     author = {Mantri, Krishna Sri Ipsit and Sasikumar, Nevasini},
     booktitle = {Medical Imaging with Deep Learning, short paper track},
     year = {2023},
     url = {https://openreview.net/forum?id=EyVbjdbQZwt},
     abbr = {MIDL},
     bibtex_show = {true},
     pdf = {https://openreview.net/pdf?id=EyVbjdbQZwt},
     html = {https://openreview.net/forum?id=EyVbjdbQZwt}
   }