1 code implementation • 12 Oct 2022 • Shiv Ram Dubey, Satish Kumar Singh, Bidyut Baran Chaudhuri
In this paper, we propose a novel AdaNorm based SGD optimizers by correcting the norm of gradient in each iteration based on the adaptive training history of gradient norm.
1 code implementation • 29 Sep 2021 • Shiv Ram Dubey, Satish Kumar Singh, Bidyut Baran Chaudhuri
The most popular and common non-linearity layers are activation functions (AFs), such as Logistic Sigmoid, Tanh, ReLU, ELU, Swish and Mish.
1 code implementation • 26 Sep 2021 • Shiv Ram Dubey, S. H. Shabbeer Basha, Satish Kumar Singh, Bidyut Baran Chaudhuri
Overall, we observe very promising performance improvement of existing optimizers with the proposed AdaInject approach.
1 code implementation • 12 Sep 2019 • Shiv Ram Dubey, Soumendu Chakraborty, Swalpa Kumar Roy, Snehasis Mukherjee, Satish Kumar Singh, Bidyut Baran Chaudhuri
In this paper, a novel optimizer is proposed based on the difference between the present and the immediate past gradient (i. e., diffGrad).
no code implementations • 23 Aug 2019 • Debabrata Paul, Bidyut Baran Chaudhuri
It sometimes recognizes a character of Bengali into the same character of a non-Bengali script, especially Assamese, which has no distinction from Bengali, except for a few characters.
Optical Character Recognition (OCR) Vocal Bursts Intensity Prediction
2 code implementations • 1 Jan 2019 • Swalpa Kumar Roy, Suvojit Manna, Shiv Ram Dubey, Bidyut Baran Chaudhuri
The proposed LiSHT is non-parametric and tackles the dying gradient problem.
no code implementations • 18 Mar 2018 • Santhosh Kelathodi Kumaran, Debi Prosad Dogra, Partha Pratim Roy, Bidyut Baran Chaudhuri
If usual traffic paths are learned in a nonparametric way, manual interventions in road marking road can be avoided.