no code implementations • 17 Aug 2023 • Sanket Jantre, Shrijita Bhattacharya, Tapabrata Maiti
In this paper, we explore two well-established shrinkage techniques, Lasso and Horseshoe, for model compression in Bayesian neural networks.
no code implementations • 15 Sep 2022 • Arkaprabha Ganguli, David Todem, Tapabrata Maiti
In recent years, deep learning has been at the center of analytics due to its impressive empirical success in analyzing complex data objects.
no code implementations • 1 Jun 2022 • Sanket Jantre, Sandeep Madireddy, Shrijita Bhattacharya, Tapabrata Maiti, Prasanna Balaprakash
Deep neural network ensembles that appeal to model diversity have been used successfully to improve predictive performance and model robustness in several applications.
no code implementations • 4 Mar 2022 • Anirban Samaddar, Sandeep Madireddy, Prasanna Balaprakash, Tapabrata Maiti, Gustavo de los Campos, Ian Fischer
In addition, it provides a mechanism for learning a joint distribution of the latent variable and the sparsity and hence can account for the complete uncertainty in the latent space.
1 code implementation • 19 Jan 2022 • Li Peide, Seyyid Emre Sofuoglu, Tapabrata Maiti, Selin Aviyente
Learning from multimodal data is of great interest in machine learning and statistics research as this offers the possibility of capturing complementary information among modalities.
no code implementations • 25 Aug 2021 • Sanket Jantre, Shrijita Bhattacharya, Tapabrata Maiti
Although several works have studied theoretical and numerical properties of sparse neural architectures, they have primarily focused on the edge selection.
1 code implementation • 26 Feb 2021 • Peide Li, Rejaul Karim, Tapabrata Maiti
Besides, we highlight the trade-off between the computational cost and the prediction risk for TEC model.
no code implementations • 19 Nov 2020 • Shrijita Bhattacharya, Zihuan Liu, Tapabrata Maiti
This paper develops a variational Bayesian neural network estimation methodology and related statistical theory.
no code implementations • 29 Jun 2020 • Shrijita Bhattacharya, Tapabrata Maiti
However there are few results which revolve around the theoretical properties of VB, especially in non-parametric problems.
1 code implementation • 28 Mar 2020 • Vojtech Kejzlar, Tapabrata Maiti
With the advancements of computer architectures, the use of computational models proliferates to solve complex problems in many scientific applications such as nuclear physics and climate research.