free web page maker

My name is Adarsh Jamadandi. I am a Masters student in Computer Science at the Saarland University (starting Nov, 2020). I completed my Bachelors in Electronics and Communication Engineering (2018) with Dr. Uma Mudenagudi as my advisor. My research interests include Graph Representation Learning and Geometric Deep Learning. I also have a deep propensity towards theoretical physics and I am looking for ways to study concepts at the intersection of theoretical physics and machine learning.

I co-founded a startup called Silicon14 (aka Tweak Labs) Inc.  during my second year of undergraduate studies. It deals with developing novel hardware augmentations for enabling rapid prototyping of embedded systems/IoT projects. I worked as a Research Assistant for Department of Science and Technology, India sponsored Underwater 3D reconstruction project at KLE Technological University.

My  side project Synoptycs.

I will be starting my Masters in Computer Science from Universit├Ąt Des Saarlandes, Saarbr├╝cken, Germany - Winter 2020/21!

Research Interests

My research interests include Graph Representation Learning and Geometric Deep Learning. More specifically, I am interested in exploring statistical/mathematical ideas that can be used to design better representation learning algorithms for graphs, texts and images. I also have a special propensity towards theoretical physics, ideas like Causal Set Theory, Tensor Networks and Renormalization Group Algorithms intrigue me. I am constantly looking at ways to work on ideas at the intersection of Machine Learning and Theoretical Physics.



Some of my creative pursuits can be found here.


* July 12 - July 16, 2021 - Excited to be part of the Geometry and Machine Learning Summer School, 2021. I will be working on Coarsening Disassortative Graphs!

* December, 2020 - Leveraging Kinematic Space for Deep Representation Learning - Accepted at NewinML, NeurIPS 2020 Workshop as Oral!

*. 2020 - Graph of Thrones : Adversarial Perturbations dismantle Aristocracy in Graphs - Accepted at AAAI'2020 Student Poster Program and the extended version is accepted at DiffGeo4DL, NeurIPS 2020.

* 2020 - Probabilistic Word Embeddings in Kinematic Space, Accepted at ICPR, 2020!

* 2019 - My interview with Hanna Ziady on bias and how researchers from Third World countries are denied VISA for attending world's largest AI conference NeurIPS - Read Story Here.

CV

Find my updated CV here.

LinkedIn

Connect with me on LinkedIn.

 ResearchGate

Follow my research on ResearchGate.

Github

Check out my Github repository.

PUBLICATIONS


1. Graph of Thrones : Adversarial Perturbations dismantle Aristocracy in Graphs. Jamadandi, Adarsh and Uma Mudenagudi. In, AAAI'2021 Student Abstract and Poster Program + Extended version at DiffGeo4DL Workshop, NeurIPS, 2020.

2. Probabilistic Word Embeddings in Kinematic Space. Jamadandi, Adarsh, Tigadoli, Rishabh, Tabib, Ramesh and Mudenagudi, Uma. In, International Conference on Pattern Recognition (ICPR), 2020.

3. Learning Hierarchical Representations in Kinematic Space.   Jamadandi, Adarsh; Tabib, Ramesh and Mudenagudi, Uma. In 33rd Conference on Neural Information Processing Systems (NeurIPS, 2019), Graph Representation Learning Workshop (Poster).

4. Exemplar based Underwater Image Enhancement augmented by Wavelet Corrected Transforms  Jamadandi, Adarsh and Mudenagudi, Uma. In the IEEE Conference on Computer Vision and Pattern Recognition,(Workshop, Oral), 2019.

5. PredGAN - A Deep Multi-Scale Video Prediction Framework for Anomaly Detection in Unlabelled Videos. Jamadandi, Adarsh;  Kotturshettar, Sunidhi and Mudenagudi, Uma In: Proceedings of 11th Indian Conference on Computer Vision, Graphics and Image Processing, 2018, ACM International Conference Proceedings (Oral).


GOOGLE SCHOLAR PROFILE

AWARDS/ACHIEVMENTS/MISCELLANEOUS 

Featured in a local Newspaper, highlighting my entrepreneurial journey!

Read the Full Story here.

My team and I designed a new line of micro-controller development boards called the Accexlron.

The board won the prestigious, "Best Practical US-Based Project" in a contest organized  by Hackster.io

Featured in TimesNext Super30 Startups!

Get in Touch!
Email : adarsh.jam@gmail.com