no code implementations • 19 Nov 2019 • Shamma Nasrin, Srikanth Ramakrishna, Theja Tulabandhula, Amit Ranjan Trivedi
To reduce the power overheads, we propose a dynamic drop out a part of the support parameters.
no code implementations • 28 Feb 2020 • Priyesh Shukla, Ahish Shylendra, Theja Tulabandhula, Amit Ranjan Trivedi
This work discusses the implementation of Markov Chain Monte Carlo (MCMC) sampling from an arbitrary Gaussian mixture model (GMM) within SRAM.
no code implementations • 25 Nov 2020 • Nick Iliev, Amit Ranjan Trivedi
We present a novel low latency CMOS hardware accelerator for fully connected (FC) layers in deep neural networks (DNNs).
no code implementations • 16 Feb 2021 • Priyesh Shukla, Ankith Muralidhar, Nick Iliev, Theja Tulabandhula, Sawyer B. Fuller, Amit Ranjan Trivedi
Addressing the computational challenges of localization in an insect-scale drone using a CIM approach, we propose a novel framework of 3D map representation using a harmonic mean of "Gaussian-like" mixture (HMGM) model.
Indoor Localization Robotics Hardware Architecture Image and Video Processing B.7; I.2.9
no code implementations • 12 Apr 2021 • Shamma Nasrin, Ahish Shylendra, Yuti Kadakia, Nick Iliev, Wilfred Gomes, Theja Tulabandhula, Amit Ranjan Trivedi
Our proposed ENOS framework allows an optimal layer-wise integration of inference operators and computing modes to achieve the desired balance of energy and accuracy.
no code implementations • 13 Nov 2021 • Priyesh Shukla, Shamma Nasrin, Nastaran Darabi, Wilfred Gomes, Amit Ranjan Trivedi
Using Bayesian inference, not only the prediction itself, but the prediction confidence can also be extracted for planning risk-aware actions.
no code implementations • 3 Mar 2023 • Alex C. Stutts, Danilo Erricolo, Theja Tulabandhula, Amit Ranjan Trivedi
Data-driven visual odometry (VO) is a critical subroutine for autonomous edge robotics, and recent progress in the field has produced highly accurate point predictions in complex environments.
no code implementations • 7 Jul 2023 • Shamma Nasrin, Maeesha Binte Hashem, Nastaran Darabi, Benjamin Parpillon, Farah Fahim, Wilfred Gomes, Amit Ranjan Trivedi
We discuss various networking configurations among CiM arrays where Flash, SA, and their hybrid digitization steps can be efficiently implemented using the proposed memory-immersed scheme.
no code implementations • 14 Jul 2023 • Davide Giacomini, Maeesha Binte Hashem, Jeremiah Suarez, Swarup Bhunia, Amit Ranjan Trivedi
Specifically, our work capitalizes on the inference mechanism of the recurrent attention model (RAM), where only a small window of input domain (glimpse) is processed in a one time step, and the outputs from multiple glimpses are combined through a hidden vector to determine the overall classification output of the problem.
no code implementations • 4 Sep 2023 • Nastaran Darabi, Maeesha Binte Hashem, Hongyi Pan, Ahmet Cetin, Wilfred Gomes, Amit Ranjan Trivedi
Moreover, our novel array micro-architecture enables adaptive stitching of cells column-wise and row-wise, thereby facilitating perfect parallelism in computations.
no code implementations • 18 Sep 2023 • Alex C. Stutts, Danilo Erricolo, Sathya Ravi, Theja Tulabandhula, Amit Ranjan Trivedi
In the expanding landscape of AI-enabled robotics, robust quantification of predictive uncertainties is of great importance.
no code implementations • 20 Sep 2023 • Domenico Parente, Nastaran Darabi, Alex C. Stutts, Theja Tulabandhula, Amit Ranjan Trivedi
This paper introduces a lightweight uncertainty estimator capable of predicting multimodal (disjoint) uncertainty bounds by integrating conformal prediction with a deep-learning regressor.
no code implementations • 11 Feb 2024 • Alex Christopher Stutts, Danilo Erricolo, Theja Tulabandhula, Amit Ranjan Trivedi
We present a novel statistical approach to incorporating uncertainty awareness in model-free distributional reinforcement learning involving quantile regression-based deep Q networks.