Publications & Books

CNN based lossy compression of hyperspectral images

Paper: Convolution neural network based lossy compression of hyperspectral images

The paper proposes a lossy hyperspectral image compression algorithm using autoencoders. It employs convolution and max-pooling layers to reduce image dimensions, generating a compressed version. Reconstruction is done using transpose convolution layers. The compressed image is entropy coded for storage or transmission. The method improves PSNR by 28% and increases compression ratio by 21 times. Classification accuracy is minimally affected, validating algorithm effectiveness.

QuantEcon[.]py

Paper: QuantEcon.py: A community based Python library for quantitative economics

Economics traditionally used simple methods for analysis, but recent trends demand computational solutions for complex, nonlinear problems. QuantEcon[.]py, an open-source Python library since 2014, supports this shift by offering a suite of tools for economic analysis. Developed for 9 years, it includes functions for numerical methods, data visualization, estimation, and dynamic programming. This article reviews its key features, showcasing its vital role in high-performance computational economics.

Dynamic Programming

Book: Dynamic Programming

This book is about dynamic programming and its applications in economics, finance, and adjacent fields like operations research. It brings together recent innovations in the theory of dynamic programming and also provides related applications and computer code.

I undertook the role of writing Python code examples and meticulously optimizing solutions for complex economic models