Md Mijanur Rahman

- Utilized Generative Adversarial Networks (GANs) to synthesize time-series data, mitigating data scarcity and addressing class imbalance challenges in deep learning contexts. - Investigated diverse loss functions and network architectures, optimizing GAN performance for robust data generation. - Compared the proposed GAN method with traditional augmentation and cutting-edge GAN techniques, showcasing innovative solutions to complex data challenges.

- Proposed the AcRCGAN model for generating synthetic time-series data, introducing a novel approach for time-series data generation. - Executed data preprocessing and established baseline classification accuracies utilizing two deep learning models on pertinent time-series datasets, ensuring robust benchmarking. - Carried out a rigorous performance evaluation of AcRCGAN compared to other state-of-the-art GANs, assessing its effectiveness and superiority.

- Proficiently conducted data wrangling on raw datasets provided by clients, ensuring data was organized and ready for analysis. - Transformed complex data into actionable insights using data visualization techniques. - Implemented demand forecasting models to boost profitability through demand prediction and optimized inventory management. - Conducted comprehensive literature reviews to remain informed about the latest advancements and best practices in demand forecasting.

Optimizing mixed-integer problems using Pyomo, a Python framework.

Run particle simulation in LAMMPS