Prabhat Kumar
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Prabhat Kumar has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
- Badri Narayan Subudhi
Indian Institute of Technology Jammu
1 shared publication - Gaetano Di Caterina
Department of Electronic and Electrical Engineering, University of Strathclyde, Glasgow, UK
1 shared publication - Himanshu Singh
Indian Institute of Technology, Jammu, Jammu, India
1 shared publication - Thangaraj Veerakumar
National Institute of Technology Goa, Cuncolim, India
1 shared publication - Vinit Jakhetiya
Indian Institute of Technology Jammu
1 shared publication
Latest Publications
- Open Access
- 17 December 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-9
Enhancing Spiking Neural Networks with Evidential Deep Learning for Object Classification on Event Based Dataset
Spiking Neural Networks (SNNs) represent a promising approach for energy-efficient, event-driven computation inspired by biological neural processes. However, their practical deployment is often hindered by three major challenges: limited availability of event-based datasets, poor interpretability of model decisions, and lack of reliable mechanisms for estimating prediction uncertainty. These limitations restrict the usability of SNNs in real-world, safety-critical applications such as robotics and edge-AI. To address these issues, this work presents a novel framework that integrates two complementary techniques: event-based data augmentation (EDA) for enhancing data diversity in neuromorphic datasets and evidential deep learning (EDL) for modeling uncertainty using a Dirichlet distribution over class evidence. We utilize a VGG-11-derived architecture, adapted into a spiking format using leaky integrate-and-fire (LIF) neurons. This architecture is trained end-to-end using surrogate gradient techniques across fixed timesteps, with temporal normalization employed to stabilize spike-based learning. An EDL module is appended to the output layer to provide calibrated confidence estimates alongside class predictions, this module is optimized using a combination of cross-entropy and KL-divergence-based evidential loss functions. We have evaluated the performance of our proposed system on two benchmark neuromorphic datasets, namely CIFAR10 DVS and N-Caltech101. Experimental results show that our proposed framework achieves classification accuracies of 76.57% and 77.20% on CIFAR10-DVS and N-Caltech101 datasets, respectively. Furthermore, we compare our technique with five state-of-the-art techniques on the aforementioned datasets, respectively, confirming the superiority of the proposed framework.
- 17 December 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-9
Open AccessEnhancing Spiking Neural Networks with Evidential Deep Learning for Object Classification on Event Based Dataset
Spiking Neural Networks (SNNs) represent a promising approach for energy-efficient, event-driven computation inspired by biological neural processes. However, their practical deployment is often hindered by three major challenges: limited availability of event-based datasets, poor interpretability of model decisions, and lack of reliable mechanisms for estimating prediction uncertainty. These limitations restrict the usability of SNNs in real-world, safety-critical applications such as robotics and edge-AI. To address these issues, this work presents a novel framework that integrates two complementary techniques: event-based data augmentation (EDA) for enhancing data diversity in neuromorphic datasets and evidential deep learning (EDL) for modeling uncertainty using a Dirichlet distribution over class evidence. We utilize a VGG-11-derived architecture, adapted into a spiking format using leaky integrate-and-fire (LIF) neurons. This architecture is trained end-to-end using surrogate gradient techniques across fixed timesteps, with temporal normalization employed to stabilize spike-based learning. An EDL module is appended to the output layer to provide calibrated confidence estimates alongside class predictions, this module is optimized using a combination of cross-entropy and KL-divergence-based evidential loss functions. We have evaluated the performance of our proposed system on two benchmark neuromorphic datasets, namely CIFAR10 DVS and N-Caltech101. Experimental results show that our proposed framework achieves classification accuracies of 76.57% and 77.20% on CIFAR10-DVS and N-Caltech101 datasets, respectively. Furthermore, we compare our technique with five state-of-the-art techniques on the aforementioned datasets, respectively, confirming the superiority of the proposed framework.