Armand Nicolicioiu

Armand Nicolicioiu

PhD Student, Machine Learning Researcher

ETH Zürich

Biography

I am a PhD student at ETH Zürich, supervised by Konrad Schindler and Răzvan Pașcanu, working in deep learning and computer vision.

I obtained my Master’s degree from University of Bucharest, doing research in out-of-distribution generalization. Previously, I obtained my Bachelor’s degree in Computer Science and Engineering at Politehnica University of Bucharest, doing research in few-shot learning as my final thesis.

During university, I did several internships: at Google Zürich,  Tübingen, and Microsoft Redmond, USA. I also worked as a Machine Learning Engineer in industry.

Download my CV.

Interests
  • Machine Learning
  • Computer Vision
  • Diffusion Models
  • Remote Sensing
Education
  • PhD in Artificial Intelligence (ongoing)

    ETH Zürich

  • MSc in Artificial Intelligence, 2022

    University of Bucharest

  • BSc in Computer Science and Engineering, 2020

    Politehnica University of Bucharest

Publications

Cheap and Effective Personalization of Foundation Language Models for Imitating a User's Writing Style
ICLR 2025 Workshop on Foundation Models in the Wild.
Neural Redshift: Random Networks are not Random Functions
CVPR 2024 (Oral Presentation)

Experience

 
 
 
 
 
Institute of Science and Technology Austria
Research Intern
Mar 2024 – Aug 2024 Klosterneuburg, Austria
  • Worked on compressed LLMs in the Distributed Algorithms and Systems Lab led by Dan Alistarh, and developed a writing assistant running entirely on-device: PanzaMail.
 
 
 
 
 
ETH Zürich
Research Intern
Oct 2023 – Jan 2024 Zürich, Switzerland
  • Research on semi-supervised learning in the Statistical Machine Learning group led by Fanny Yang.
 
 
 
 
 
Idiap Research Institute
Research Intern
May 2023 – Sep 2023 Martigny, Switzerland
  • Research on the robustness of diversified model ensembles and on the generalization properties of neural networks. Supervisor: Damien Teney.
 
 
 
 
 
Amazon
SDE Intern
Nov 2022 – Mar 2023 Tübingen, Germany
  • Improved an evaluation framework for assaying out-of-distribution generalization in the Causal Representation Learning team.
 
 
 
 
 
Google
Software Engineering Intern
Jun 2022 – Sep 2022 Zürich, Switzerland
  • Used large language models (LLMs) to process the transcripts of YT videos for entity linking.
  • Generated realistic synthetic data to complement existing training sets.
 
 
 
 
 
dotLumen
Machine Learning Researcher
Oct 2020 – Nov 2021 Cluj-Napoca, Romania (Remote)
  • Prototyped a headset to help blind persons be more independent..
  • Multi-modal environment understanding using Computer Vision.
 
 
 
 
 
Microsoft
Software Engineering Intern
Jul 2019 – Sep 2019 Redmond, WA, USA
  • Designed and built a fast generic trending system for large scale data.
  • Microservices pipeline for ingestion, processing, storing and retrieval.
 
 
 
 
 
Microsoft
Software Engineering Intern
Jul 2018 – Sep 2018 Redmond, WA, USA
  • Improved Azure’s real-time telemetry monitoring system by generating recommendations that lead to a faster resolution of alerts.
 
 
 
 
 
Arnia Software
Machine Learning and Computer Vision Junior Researcher
Jan 2017 – Nov 2017 Bucharest, Romania
  • Developed algorithms for improving the HDR camera mode of phones.
  • Researched generic object segmentation and color perception.
 
 
 
 
 
Sparktech Software
Machine Learning Engineer Intern
Jul 2016 – Oct 2016 Bucharest, Romania
  • Built a recommender system for a coupon site from scratch.

Honors & Awards

  • Poster Award & Scholarship at Synapse AI Symposium 2023 by Bending Spoons in Milan for my research project on model diversity.

  • Best Poster Award at Romanian AI Days 2022 conference for presenting my Master’s thesis on Diversifying Vision Transformers for Out-of-Distribution Generalization.

  • Travel Scholarship at Synapse AI Symposium by Bending Spoons in Milan
  • 1st place at the Computational Geometry & Topology Challenge at ICLR 2021 for our Study on Topological Noise Invariant Features.
  • Best Poster Award at Eastern European Machine Learning Summer School (EEML 2020) for presenting my Bachelor’s thesis on Few shot learning by features adaptation with Graph Neural Networks.
  • 1st place at Catalyst Coding Contest 2019.
  • 1st place at Textract 2018.
  • 1st place at ITFest 2017.
  • Qualified for the ACM-ICPC Southeastern European Regional Contest (SEERC 2016) as a 1st year student.
  • Romanian National Olympiad of Informatics:

    • Bronze Medal: 2013, 2015
    • Qualified / Participated: 2012, 2014
  • Romanian National Olympiad of Physics:

    • Bronze Medal: 2012
    • Participated: 2010, 2011
  • Romanian National Olympiad of Mathematics:

    • Qualified / Participated: 2009, 2012