Mr. Rupam K. Karmakar

Computer Science Researcher & Machine Learning Specialist

Welcome to my academic portfolio. I am a computer science researcher specializing in machine learning and distributed systems. My work focuses on developing innovative algorithms and systems that push the boundaries of large-scale computation and artificial intelligence.

15+
Research Projects
8
Publications
500+
Citations
4
Years Experience

Institution: Department of Computer Science, University Name

GitHub: github.com/johndoe

LinkedIn: linkedin.com/in/johndoe

Google Scholar: View Profile

About Me

I am a doctoral candidate in Computer Science at University Name, where I work under the supervision of Professor Jane Smith. My research focuses on the intersection of machine learning and distributed systems, particularly on developing efficient algorithms for large-scale data processing and neural network training.

My work has been published in top-tier conferences including ICML, NeurIPS, and ICLR. I am passionate about bridging the gap between theoretical research and practical applications, with a focus on making advanced machine learning techniques accessible and efficient.

Research Interests

Machine Learning Distributed Systems Algorithm Design High-Performance Computing Neural Network Optimization Deep Learning Graph Processing

Education

Ph.D. in Computer Science

University Name, 2021 – Present

Dissertation: "Efficient Training Algorithms for Large-Scale Neural Networks in Distributed Environments"

Advisor: Professor Jane Smith

M.S. in Computer Science

Another University, 2018 – 2020

Thesis: "Neural Network Optimization Techniques for Resource-Constrained Devices"

GPA: 3.9/4.0

B.E. in Computer Engineering

Institute Name, 2014 – 2018

Honors: Summa Cum Laude, Dean's List (all semesters)

GPA: 3.95/4.0

Awards & Honors

Research Projects

Distributed Training Framework for Neural Networks

2023 – Present | Lead Researcher

Developed a novel distributed training framework that reduces training time for large neural networks by 40% through innovative gradient aggregation techniques. The framework supports heterogeneous computing clusters and provides fault tolerance mechanisms, making it suitable for both academic research and industrial applications.

Technologies:

Python C++ MPI CUDA TensorFlow PyTorch
Efficient Data Partitioning for Graph Algorithms

2022 – 2023 | Co-Investigator

Implemented and evaluated various graph partitioning strategies for distributed graph processing systems. Achieved 2x speedup on benchmark datasets compared to existing approaches. The research addresses fundamental challenges in load balancing and communication overhead in distributed graph computations.

Technologies:

C++ Apache Spark GraphX Hadoop
Neural Architecture Search Optimization

2021 – 2022 | Principal Investigator

Designed an automated neural architecture search system using evolutionary algorithms and reinforcement learning. The system discovers optimal network architectures for specific tasks while maintaining computational efficiency. Successfully deployed in production environments for image classification and natural language processing tasks.

Technologies:

Python PyTorch NumPy SciPy
Memory-Efficient Training Algorithms

2020 – 2021 | Research Assistant

Researched and implemented memory-efficient training techniques for deep learning models, enabling the training of larger models on resource-constrained hardware. Developed gradient checkpointing strategies that reduce memory consumption by 60% with minimal impact on training speed.

Technologies:

Python TensorFlow Keras CUDA

No projects found matching your search.

Publications

Efficient Gradient Aggregation in Distributed Neural Network Training

J. Doe, J. Smith, A. Johnson

International Conference on Machine Learning (ICML), 2024 | Best Paper Award

This paper presents a novel approach to gradient aggregation in distributed training environments. Our method reduces communication overhead by 35% while maintaining convergence properties. The technique has been successfully deployed in production systems handling petabyte-scale datasets.

[PDF] [Code] [Slides]

Graph Partitioning Strategies for Large-Scale Distributed Computing

J. Doe, K. Lee

ACM Symposium on Parallel Algorithms and Architectures (SPAA), 2023

We analyze different graph partitioning strategies and propose a hybrid approach that adapts to graph topology characteristics, resulting in improved load balancing. Our method shows consistent improvements across various real-world graph datasets including social networks and web graphs.

[PDF] [Code]

Scalable Training Infrastructure for Foundation Models

J. Doe, M. Chen, R. Patel, J. Smith

Conference on Neural Information Processing Systems (NeurIPS), 2023

We present a comprehensive infrastructure solution for training foundation models at scale. Our system handles model parallelism, pipeline parallelism, and data parallelism efficiently, enabling the training of models with over 100 billion parameters on clusters with thousands of GPUs.

[PDF] [arXiv]

Automated Neural Architecture Design Using Evolutionary Algorithms

J. Doe, M. Chen, J. Smith

Neural Information Processing Systems (NeurIPS), 2022

This work introduces an evolutionary algorithm framework for automated neural architecture search that discovers architectures competitive with hand-designed networks. Our approach reduces the computational cost of architecture search by an order of magnitude compared to previous methods.

[PDF] [Code] [Poster]

Memory-Efficient Training for Deep Neural Networks

J. Doe, R. Brown

International Conference on Learning Representations (ICLR), 2021

We propose a memory optimization technique that enables training of deep networks with limited GPU memory through gradient checkpointing and dynamic memory allocation. Our approach achieves a 60% reduction in memory usage while maintaining training efficiency within 5% of the baseline.

[PDF] [Code]

No publications found matching your search.

Notebooks

Efficient Gradient Aggregation in Distributed Neural Network Training

J. Doe, J. Smith, A. Johnson

International Conference on Machine Learning (ICML), 2024 | Best Paper Award

This paper presents a novel approach to gradient aggregation in distributed training environments. Our method reduces communication overhead by 35% while maintaining convergence properties. The technique has been successfully deployed in production systems handling petabyte-scale datasets.

[PDF] [Code] [Slides]

Graph Partitioning Strategies for Large-Scale Distributed Computing

J. Doe, K. Lee

ACM Symposium on Parallel Algorithms and Architectures (SPAA), 2023

We analyze different graph partitioning strategies and propose a hybrid approach that adapts to graph topology characteristics, resulting in improved load balancing. Our method shows consistent improvements across various real-world graph datasets including social networks and web graphs.

[PDF] [Code]

Scalable Training Infrastructure for Foundation Models

J. Doe, M. Chen, R. Patel, J. Smith

Conference on Neural Information Processing Systems (NeurIPS), 2023

We present a comprehensive infrastructure solution for training foundation models at scale. Our system handles model parallelism, pipeline parallelism, and data parallelism efficiently, enabling the training of models with over 100 billion parameters on clusters with thousands of GPUs.

[PDF] [arXiv]

Automated Neural Architecture Design Using Evolutionary Algorithms

J. Doe, M. Chen, J. Smith

Neural Information Processing Systems (NeurIPS), 2022

This work introduces an evolutionary algorithm framework for automated neural architecture search that discovers architectures competitive with hand-designed networks. Our approach reduces the computational cost of architecture search by an order of magnitude compared to previous methods.

[PDF] [Code] [Poster]

Memory-Efficient Training for Deep Neural Networks

J. Doe, R. Brown

International Conference on Learning Representations (ICLR), 2021

We propose a memory optimization technique that enables training of deep networks with limited GPU memory through gradient checkpointing and dynamic memory allocation. Our approach achieves a 60% reduction in memory usage while maintaining training efficiency within 5% of the baseline.

[PDF] [Code]

No Notes found matching your search.

Skills & Expertise

Programming Languages

Python C++ Java CUDA JavaScript SQL R Bash

Machine Learning Frameworks

TensorFlow PyTorch Keras Scikit-learn JAX Hugging Face XGBoost

Distributed Computing

Apache Spark Hadoop MPI Ray Dask Kubernetes Docker

Cloud Platforms

AWS Google Cloud Azure Lambda Labs

Research Skills

Algorithm Design Performance Optimization Experimental Design Statistical Analysis Technical Writing Presentation Mentoring

Tools & Technologies

Git Linux Jupyter LaTeX Matplotlib Weights & Biases TensorBoard