Hey! I'm Ramith πŸ˜ƒ

Ramith, smiling and waving at the camera

"Maybe so, maybe not, we'll see.."

Email  /  CV  /  Google Scholar  /  Github

I'm a Post-Bac Fellow at Harvard, and a Visiting Scientist at MIT working at SoLab.
I've been engaging in computational biology and imaging research with the guidance of
Dr. Dushan Wadduwage and Dr. Sergey Ovchinnikov.

>> I'm interested in developing algorithms and fusing insights from machine learning (ML) to advance biology & healthcare, while striving for robustness and interpretability <<

I started this blog when I was learning Signals and Systems, and through this blog,
I aspire to clearly write (+ develop visualizations)$^\dagger$ about concepts & advancements that awe me.
Occasionally I record piano covers and try to sing songs. You can find them here. 🎢🎹

Research Interests:Computational Biology,Interpretable ML Models for Biology, ML For Science
Multi-modal Learning (Genomics+Imaging)
Research Directions:Differentiable Algorithms, Combinatorial Optimization,Graph Theory/Representation Learning
Uncertainty Quantification

University of Moratuwa, Sri Lanka
Oct 2017 - June 2022

CSIRO | Data61
Oct 2020 - May 2021

Harvard University
July 2022 - Present

MIT
Jan 2024 - Present

Project AyaοΉ‘
June 2023 - Present

οΉ‘An open science project

$\dagger$ Inspired by Distill.pub

News

2023 June
πŸ“‘ Papers accepted to workshops, Sampling and Optimization in Discrete Space (SODS) ፨ and Differentiable Almost Everything (DiffAE) 〆 at ICML 2023 πŸŽ‰
2022 November
[Invited Talk] - Northeast Symposium on Biomedical Optics
2022 September
[Invited talk] - Nano-SymBioSys workshop at UiT, The Arctic University of Norway πŸ‡³πŸ‡΄
2022 August
πŸ“„ Paper accepted to IEEE Signal Processing Magazine
2022 July
  
Started work as a Post Baccalaureate Fellow at Harvard
2021 June
πŸ… Our team - $T^3$, became the Champions of the IEEE Signal Processing Cup 2021 πŸŽ‰
2021 February
πŸ“„ First Author Paper Accepted to ISCAS 2021
2019 October
πŸ… Our team, became 48th in the world, 1st in πŸ‡±πŸ‡° in IEEEXtreme 13.0 Competition

Want to chat?


I allocate ~1.5 hours per week to have 1:1 sessions with undergrads who wants any advice/guidance. If you are such an undergrad, you can schedule a time here.
If the link doesn't seem to work, please try in incognito mode.

24 Feb 13

An inspiring open science journey to remember πŸ’™

5 min read
22 Mar 08

Access Harvard FASRC through vscode without entering password & two step code every time

1 min read
22 Feb 25

How to setup a JAX/Tensorflow 1.15 environment in the FASRC Cluster

5 min read
22 Jan 27

Deep Residual Learning for Image Recognition

3 min read
22 Jan 05

JAX 001 - A closer look at its background

3 min read
21 Sep 27

Summary of Paper - Representation Learning Via Invariant Causal Mechanisms (ReLIC)

2 min read
19 Jul 19

Learning a new hardware platform (ScholarX - Part 2)

2 min read
19 Jul 14

Making a robot Visualisation Platform from Scratch (and how you can do the same)

2 min read
19 Jun 20

What I learnt the hard-way | Taking notes

1 min read
19 Jun 03

Building Blocks Matter!: #ScholarX

3 min read
18 Dec 28

Creating a pulse sensor from scratch πŸ’Ÿ

1 min read
18 Dec 22

Creating a sustainable blog..? Β―\_(ツ)_/Β―

2 min read
18 Dec 04

Analysing Second Order System Response of an RLC Circuit

1 min read
18 Dec 02

Does Bilateral Laplace Transform Exist for Sin(t) ?

1 min read
18 Dec 02

Understanding Properties of the Laplace Transform using Wolfram Language 🐺

2 min read




Some things I care about (or used to do) during my free time...
Project Aya
Open Science Effort to Accelerate Multilingual AI

Project Aya - Sinhala/ΰ·ƒΰ·’ΰΆ‚ΰ·„ΰΆ½

Working alongside likeminded Sinhala speaking individuals who'd love to see our language thrive in the current NLP advancement landscape.
(2023 - Present)
Audiobook library

Manusath Handa

We built on online platform to gather and disseminate audiobooks (Sinhala/Tamil) to visually impaired students.
(Contributed while I was an Undergrad).
Some Visualization Tools I've built
D* Lite
Algorithm
Visualizer

D* Lite Algorithm

As part of my internship, I wanted to implement the D* Lite algorithm, with dynamic obstacle creation aspect. (WebAssembly)
Intelliscope
Convolutional NN Visualizer

Project Intelliscope

As part of a medical research, built a model visualization tool to get more intuition on what sort of learning has the model performed. (TensorSpace.js)
Multi-Core Processor
Simulator

FPGA Group Project

To test our ISA and the functionality of any assembly code, I developed a tool to simulate the activity of each core. (p5.js)