Apoorva Vikram Singh

Apoorva Vikram Singh

Masters Student in Neural Information Processing

Eberhard Karls University of Tübingen, Germany


I am a first year master student in the Graduate Training Centre of Neuroscience at the Eberhard Karls University of Tübingen, Germany.

I am deeply fascinated by Machine learning and Deep Learning and the underlying high dimensional mathematical modeling that enables it. The level of generalization and adaptability in deep learning architectures needs to attain a certain level before they can be used as decision-making models instead of prediction models. The introduction of simple perturbations on data distributions can render the best of the deep learning models ineffective. This raises concerns on the wide variety of learning theories that form the core of modern machine and deep learning architectures. My long term vision is to revisit the fundamental theory of deep learning to understand when and why they fail.

  • Theoretical Machine Learning
  • Computer Vision
  • Applied Deep Learning
  • M.S. in Neural Information Processing, 2021-2023

    Eberhard Karls University of Tübingen, Germany

  • B.Tech in Electrical Engineering, 2017-2021

    National Institute Of Technology, Silchar

  • Higher Secondary Education (ISC), 2014-2017

    City Montessori School, Lucknow


Student Developer
Jun 2021 – Aug 2021 Rhode Island, U.S.A
  • Worked on the project Equivariant Neural Networks for Dark Matter Morphology with Strong Gravitational Lensing under Prof. Sergei Gleyzer and Michael Toomey.
Research Intern
Dec 2020 – Dec 2021 Rhode Island, U.S.A
Guest Researcher
Jun 2020 – Feb 2021 Magdeburg, Germany
  • Working under Prof. Peter Benner and Dr. Pawan Goyal in the research group Computational Methods in Systems and Control Theory.
  • Working on methods of inverse imaging problems using deep learning.
Undergraduate Research Intern
May 2019 – Jul 2019 Hyderabad, India
  • Worked in Artificial Intelligence (AI) Laboratory, UoH, under Prof. Atul Negi, School of Computer and Information Sciences.
  • Worked on developing efficient drug repositioning techniques by using ontological medical data and medical text corpus.
Machine Learning Engineer
Dec 2017 – Aug 2018 Silchar, India
  • Worked to develop a deep learning based engine that analysed the client’s monthly health data to anticipate any health risks.
  • The data was collected through wearables and uploaded on monthly basis to a cloud server.

Recent Publications

Quickly discover relevant content by filtering publications.
(2021). DynG2G: An Efficient Stochastic Graph Embedding Method for Temporal Graphs. arXiv preprint arXiv:2109.13441.

PDF Cite

(2020). A Hybrid Classification Approach using Topic Modeling and Graph Convolution Networks. 2020 International Conference on Computational Performance Evaluation (ComPE).

PDF Cite Code Slides

(2020). Debunking Fake News by Leveraging Speaker Credibility and BERT Based Model. IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology (WI-IAT'20).

PDF Cite Code Slides

(2020). Predictive approaches for the UNIX command line: curating and exploiting domain knowledge in semantics deficit data. Multimedia Tools and Applications.

PDF Cite Code

(2020). Seq2Seq and Joint Learning Based Unix Command Line Prediction System. arXiv preprint arXiv:2006.11558.

PDF Cite Code