Skip to content
View TheAlphaJas's full-sized avatar

Block or report TheAlphaJas

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
TheAlphaJas/README.md

Profile Views

Hey there! 👋 I'm Jasmer

Personal website revamp in progress — beta version here(may not be fully updated)! 🚧

LinkedIn Codeforces Google Scholar Email

Welcome to my GitHub profile! B.Tech undergrad at Indian Institute of Technology (IIT) Indore. I like machine learning systems, causal reasoning, signal processing, and building things from scratch.

Currently/Previously worked on :

  • Causal Reinforcement Learning
  • Algorithmic Decision Making
  • Communication Systems & 6G
  • Speech & Audio Processing
  • Medical Imaging

CV - If Interested!


Current / Recent Work

Causal Reinforcement Learning (Dr. Dibbendu Roy, Ongoing Research)

  • Exploring correlation vs causation in sequential decision-making settings.
  • Studying structural causal models (SCMs) for RL environments.
  • Strong theoretical + algorithmic focus (not just empirical RL).
  • Check out final slides here. Final report here

Speech Synthesis with In-The-Wild (ITW) Data (LIPG,IITI & RUG-NL(Campus Fryslân), Ongoing Research)

  • Studying TTS systems trained on noisy, unconstrained real-world speech.
  • Investigating evaluation gaps between traditional metrics (e.g., DNSMOS/UTMOS) and perceptual quality in ITW setups.
  • Goal: bridge controlled lab-quality TTS and messy real-world deployment.

6G & CSI-RS (Samsung R&D Institute India – 2025)

  • Worked with the 6G Standards Team on ML-based CSI-RS processing.

Audio-Visual Speech Enhancement (OPTIMAL Lab, IIT Indore, Ongoing Research)

  • Developed LSTMSE-Net (accepted at InterspeechW 2024 – AVSEC).
  • Engineered temporal feature extraction pipelines using RNN/LSTM.
  • Next version in current development

Medical Image Segmentation (LIPG, IIT Indore & Northwestern U)

  • Built advanced U-Net variants for liver & tumor segmentation.
  • Integrated SE blocks, ASPP modules.
  • Co-authored multiple papers in:
    • Biomedical Signal Processing and Control (Elsevier)
    • Frontiers in Medicine (2025)

Current Learning Directions

  • Causal Inference for Sequential Decision Making
  • Robust & Distribution-Shift-Aware RL
  • Advanced Communication & Information Theory
  • Convex Optimization
  • Elementary Measure Theory, Probability Theory and Statistics
  • Group Theory (Welcome Abstract Algebra!)
  • Game Theory & Mechanism Design
  • Misc - Competitve Programming when I get the time!

📈 Stats



"Every algorithm is a small step towards defining and understanding intelligence."

Pinned Loading

  1. Liver-Tumor-Seg-Implements Liver-Tumor-Seg-Implements Public

    This repository contains codes of various state-of-the-art methods and research papers for Liver Tumor Segmentation

    Jupyter Notebook 7

  2. rawML-Python rawML-Python Public

    Neural Networks without any high level API.

    Jupyter Notebook 2

  3. HiFormer-LITS HiFormer-LITS Public

    Forked from amirhossein-kz/HiFormer

    Implementation of the LITS dataset on HiFormer architecture

    Python 4

  4. encrypgen2 encrypgen2 Public

    Text encrypter gen2. Encrypts a single line of input into a file. Coded in Python, but the algorithm can be theoretically coded in any language consisting lists and maps (dictionarys).

    Python 2

  5. G-CASCADE-LiTS G-CASCADE-LiTS Public

    Forked from SLDGroup/G-CASCADE

    Implementation of G-CASCADE on the LITS Dataset

    Python 3

  6. Striver-TuF-A2Z Striver-TuF-A2Z Public

    This repo contains my solutions to Striver's A2Z DSA sheet

    C++ 1