Md Tarekul Islam

Accomplished ML Research Associate with a proven track record at the Technical University of Dresden, adept in PyTorch, Tensorflow and CUDA. Proficient in utilizing various machine learning algorithms, deep learning frameworks and data manipulation techniques to derive insights from complex datasets. Demonstrated exceptional problem-solving skills and a commitment to innovation, significantly improving machine learning methods, and presenting findings to key stakeholders.

Title – “Possibilities of Applying Machine learning Techniques for probabilistic Analysis in a Turbine Blade-Disk Interface.”

Conference: Proceedings of the ASME Turbo Expo 2024, Turbomachinery Technical Conference and Exposition (GT2024)

Paper ID: GT2024-12247

Location: London, United Kingdom

PyTorch for Deep learning and machine learning full course from Free Code camp

• Contributed to a team project focused on generating 3D objects using Generative Adversarial Networks (GANs) by leveraging high performance GPU computing systems for efficient training and inference.
• Implemented four different methods for Natural Language Processing (NLP): TF-IDF, Word2Vec, Doc2Vec, and Recurrent Neural Network (RNN) utilized for a specific NLP task, and their performances are compared using precision, recall, accuracy, and other relevant metrics.
• Instance segmentation using the YOLOv8 model, as well as developed algorithms for converting masks to polygons for both single and multi-instance segmentation.
• Forecasted carbon monoxide concentration for a particular geospatial location for a specific month using different station data.