Profile image Profile image

Taposh Dutta Roy

Manager, Innovation Team

Cellular Monte Carlo Modeling of Al (x) In (1-x) Sb/InSb Quantum Well Transistors

In this work, an Indium Antimonide (InSb) quantum well transistor is investigated using full-band Monte Carlo simulations. The steady-state characteristic of the device is first analyzed, showing particle transport along the two-dimensional electron gas (2DEG). The small-signal behavior of the device is also investigated. Finally, the noise analysis is performed, allowing for a two-dimensional mapping of the noise within the device.

Paper Download
Cellular Monte Carlo Modeling

In this “Poster presentation” simulation of quantum well transistors using full-band Monte Carlo simulation is explained.


Division algorithms have been developed to reduce latency and to improve the efficiency of the processors. Floating point division is considered as a high latency operation. This papers looks into one such division algorithm, examines the hardware block diagram and suggests an alternative path which may be cost effective.

Paper Download


Open Source libarary for forecasting in Python - this has implementation of all traditional methods and newer ones such as - 1. Theta Method 2. N-BEATS: Neural basis expansion analysis for interpretable time series forecasting.

AutoML and interpretability in healthcare

The healthcare industry requires accuracy and highly interpretable models. Depending on the application domain in healthcare (clinical versus operational), the models can range from forecasting to segmentation and clustering to classification and regression, including areas like natural language processing (NLP) and image processing. However, the data is usually plagued by missing information and incorrect values, so data preparation and devel... read more

A hybrid approach to model randomness and fuzziness using Dempster-Shaffer method.

Current main stream data science work is mainly focused on prediction , segmentation and data analysis. This mainly involves supervised learning where we learn from historic data. The predicted observations usually gives a probability score which measures randomness. We believe there are situations where along with randomness we need some fuzziness to give some confidence to the observation. Fuzziness and randomness are two very different conc... read more


Basics of GPU Computing for Data Scientists

GPU’s have become the new core for image analytics. More and more data scientists are looking into using GPU for image processing. In this article I review the basics of GPU’s that are needed for a data scientist and list a frame work discussed in literature for suitability of GPU for an algorithm. Lets start with what is a GPU? Graphics Processing Unit (GPU) were originally created for rendering graphics. However due to their high performan... read more


Machine Learning to Save Lives

Talk in 2015 on how to use ML to save lives. H2O World 2015, Day 2 Join the Movement: open source machine learning software from, go to Github repository Do you like this? Check out more talks on open source machine learning software at:

Healthcare Panel - Change Healthcare, Kaiser Permanente, Sanofi, Stanford University

This panel was recorded at #H2OWorld 2017 in Mountain View, CA. Learn more about Follow @h2oai: - - - Moderated by: Sanjay Joshi, CTO, Healthcare & Life Sciences, Panelists: - Taposh Dutta Roy, Thought Leader, Kaiser Permanente - Adam Sullivan, Senior Director, AI, Change Healthcare - Somalee Datta, Director of Research IT, Stanford University - Woranat Wongdhamm... read more