prof_pic.jpg

Md Awsafur Rahman

PhD Student at UC Santa Barbara | Ex - Google | Kaggle Grandmaster & Google Developer Expert (GDE) | Ex - Weights & Biases | Ex - Research Assistant in IRAB

Hello! I’m a PhD student in the ECE department at University of California, Santa Barbara (UCSB). Beyond that, I am a Kaggle Grandmaster and Google Developer Expert (GDE). Recently, I left my job (contract) for Google, where I worked on solving competitive Machine Learning problems from Kaggle in the fields of Computer Vision, Natural Language Processing, and Audio Processing, utilizing multi-backend Keras.

Just like a detective, I find joy in dissecting complex research problems and crafting solutions through code, and math. My perseverance has led to the authorship of 7 papers and co-authorship of 2 more. Beyond academia, I thrive in competitions like Kaggle, DrivenData, etc. It’s here that I became one of the youngest Kaggle Grandmasters, peaking at 5th (now 8th) out of 61,000+ participants. I also take pride in open-source contributing on GitHub, in platforms such as TensorFlow, Keras, HuggingFace, YOLOv5 (PyTorch), etc. I also won Google OSS Prize 4 times and Kaggle ML Research Award for my open-source contributions.

Interests

  • Multi-modal Learning
  • Generative AI
  • Vison-Language Model
  • Deepfake Detection

Education

  • PhD in Electrical & Computer Engineering (2024 - Present)
    University of California, Santa Barbara (UCSB), USA
  • BSc in Electrical & Electronic Engineering (2018 - 2023)
    Bangladesh University of Engineering and Technology (BUET), Bangladesh

News

Dec 1, 2023 Started working for Google in the Keras Team as a Contractor.
Oct 9, 2023 Awarded Google Open Source Peer Bonus Award for gcvit-tf and TransUNet-tf.
Aug 15, 2023 Symbiotic Transformer paper has been accepted to WACV 2024
Jul 17, 2023 DwinFormer paper has been accepted to IEEE Sensors Journal (Q1, IF: 4.3)
Jun 22, 2023 ArtiFact paper has been accepted at ICIP 2023
Jun 1, 2023 Joined IRAB (Institue of Robotics & Automation BUET) as Research Assistant.
Sep 10, 2022 Ranked 1st on LB in IEEE VIP Cup 2022: Synthetic Image Detection at ICIP 2022.
Sep 3, 2022 Won Kaggle ML Research Spotlight: August 2022 Award for GCViT-TF.
Jul 11, 2022 Won Google OSS Expert Prize: June 2022 Award for Fake Speech Detection.
Jul 3, 2022 Ranked 1st in IEEE SP Cup 2022: Synthetic Speech Attribution at ICASSP 2022.

Selected Publications

2024

  1. /assets/paper/syn-tra/animation.gif
    Semi-Supervised Semantic Depth Estimation using Symbiotic Transformer and NearFarMix Augmentation
    Md Awsafur Rahman, and Shaikh Anowarul Fattah
    WACV, 2024
  2. /assets/paper/syn-att/animation.gif
    Syn-Att: Synthetic Speech Attribution via Semi-Supervised Unknown Multi-Class Ensemble of CNNs
    Md Awsafur Rahman, Bishmoy Paul, Najibul Haque Sarker, and 3 more authors
    2024
  3. /assets/paper/sonics/sonics.gif
    SONICS: Synthetic Or Not - Identifying Counterfeit Songs
    Md Awsafur Rahman, Zaber Ibn Abdul Hakim, Najibul Haque Sarker, and 2 more authors
    2024

2023

  1. /assets/paper/dwinformer/animation.gif
    DwinFormer: Dual Window Transformers for End-to-End Monocular Depth Estimation
    Md Awsafur Rahman, and Shaikh Anowarul Fattah
    IEEE Sensors Journal, 2023
  2. /assets/paper/artifact/animation.gif
    ArtiFact: A Large-Scale Dataset with Artificial and Factual Images for Generalizable and Robust Synthetic Image Detection
    Md Awsafur Rahman, Bishmoy Paul, Najibul Haque Sarker, and 2 more authors
    ICIP, 2023
  3. /assets/paper/ciff-net/animation.gif
    CIFF-Net: Contextual Image Feature Fusion for Melanoma Diagnosis
    Md Awsafur Rahman, Bishmoy Paul, Tanvir Mahmud, and 1 more author
    Elsevier BSPC, 2023

2022

  1. /assets/paper/embc22/animation.gif
    A Deep Learning Scheme for Detecting Atrial Fibrillation Based on Fusion of Raw and Discrete Wavelet Transformed ECG Features
    Md Awsafur Rahman, Shahed Ahmed, and Shaikh Anowarul Fattah
    In EMBC, 2022

2021

  1. /assets/paper/covsegnet/animation.gif
    CovSegNet: A multi encoder–decoder architecture for improved lesion segmentation of COVID-19 chest CT scans
    Tanvir Mahmud, Md Awsafur Rahman, Shaikh Anowarul Fattah, and 1 more author
    IEEE Transactions on Artificial Intelligence, 2021

2020

  1. /assets/paper/covxnet/animation.gif
    CovXNet: A multi-dilation convolutional neural network for automatic COVID-19 and other pneumonia detection from chest X-ray images with transferable multi-receptive feature optimization
    Tanvir Mahmud, Md Awsafur Rahman, and Shaikh Anowarul Fattah
    Computers in biology and medicine, 2020