---

I am a research scientist at Nokia Bell Labs and a visiting researcher at the University of Cambridge. My work enables AI to make the most out of real-world data (e.g. timeseries, audio, mobile sensors, or other modalities), through data-efficient and robust models. I am particularly interested in the following paradigms:

• Self-supervision. Contrastive, masked, or generative - what comes after that?
• Transfer learning. How can models generalize out of distribution - what are their limits?
• Multi-modality/tasking. How to merge modalities in a data-agnostic way?

Previously, I completed a PhD in Computer Science at the University of Cambridge working with Prof. Cecilia Mascolo. During my studies, I was fortunate to work in diverse industries and companies including Microsoft Research, Telefonica Research, Ocado, Qustodio (acquired for $50M), and research labs.

My research has been published in top venues in artificial intelligence, ML for health, human-computer interaction, and signal processing. Further, I am part of COVID-19 Sounds, an audio-AI study which builds predictive models for COVID-19 using smartphone respiratory recordings. Recent projects have been featured in international media such as the BBC, CNN, Forbes, Financial Times, and Guardian (see more below).

CV

📖 Publications

---

2022

2021

2020

2019

Pre-PhD (2013-2018)

Interactive dimensionality reduction using similarity projections
Dimitris Spathis, Nikolaos Passalis, Anastasios Tefas
Knowledge-Based Systems, 165

Fast, Visual and Interactive Semi-supervised Dimensionality Reduction
Dimitris Spathis, Nikolaos Passalis, Anastasios Tefas
ECCV Efficient Feature Representation Learning workshop (CEFRL @ ECCV 2018), Munich, Germany

Diagnosing Asthma and Chronic Obstructive Pulmonary Disease with Machine Learning
Dimitris Spathis, Panayiotis Vlamos
Health Informatics Journal, 25(3)

Class-based Prediction Errors to Detect Hate Speech with Out-of-vocabulary Words
Joan Serra, Ilias Leontiadis, Dimitris Spathis, Gianluca Stringhini, Jeremy Blackburn, Athena Vakali
ACL Abusive Language Online workshop (ALW @ ACL 2017), Vancouver, Canada

A comparison between semi-supervised and supervised text mining techniques on detecting irony in greek political tweets
Basilis Charalampakis, Dimitris Spathis, Elias Kouslis, Katia Kermanidis
Engineering Applications of Artificial Intelligence, 51

Detecting Irony on Greek Political Tweets: A Text Mining Approach
Basilis Charalampakis, Dimitris Spathis, Elias Kouslis, Katia Kermanidis
International Conference on Engineering Applications of Neural Networks, Rhodes, Greece

Glocal News: An Attempt to Visualize the Discovery of Localized Top Local News, Globally
Dimitris Spathis, Theofilos Mouratidis, Spyros Sioutas, Athanasios Tsakalidis
International Conference on Conceptual Modeling, Hong Kong, China

Theses

---

Machine learning to model health with multimodal mobile sensor data
PhD thesis
University of Cambridge, 2021

Learning to interact with high-dimensional data
MSc thesis
Aristotle University, 2017

🧐 Academic service

---

Organizer:

• Co-organizer and track chair of CHIL 2023, Boston, USA.
• Senior panel/roundtable chair at ML4H 2022, New Orleans, USA.
• Chair of WellComp 2022 workshop at UbiComp 2022, Cambridge, UK.
• Co-organizer of the Federated sensing tutorial at MobiCom 2021, New Orleans, USA.

Program Committee Member: AAAI 2021-2023, IJCAI 2020, KDD 2020-2023 (PC & Session Chair), SIAM SDM 2022, Sensiblend @ Ubicomp 2021, Mobiquitous 2022.

Reviewer: NeurIPS, ICLR, ICML, AAAI, IJCAI, KDD, CHI, Ubicomp/IMWUT, CHIL, Nature Digital Medicine, Nature Scientific Reports, ICASSP, Expert Systems with Applications, Neurocomputing, WWW/The Web Conference, Engineering Applications of Artificial Intelligence, Sensors, ICWSM, ICPR, Pervasive and Mobile Computing, Computer Networks, The Visual Computer, and more.

📢 Talks

---

Looking for Out-of-Distribution Environments in Multi-center Critical Care Data

• 📍 ML4H'22, New Orleans, USA (co-located with NeurIPS'22) — November 28, 2022

Representation learning for cardio-fitness prediction in free-living environments

• 📍 King's College London, Precision Health Informatics Data Lab, London, UK — November 24, 2022

AI-powered Wearables Transforming Mobile Health

• 📍 AI Summit, London Tech week, London, UK — June 16, 2022

Federated learning for human activity recognition

• 📍 Mobicom'21 (held 2022), New Orleans, USA — April 1, 2022

Self-supervised learning for health signals

• 📍 Feinstein Institutes of Northwell Health, New York, USA (remote) — March 22, 2022

COVID-19 Sounds: A Large-Scale Audio Dataset for Digital Respiratory Screening

• 📍 NeurIPS'21, virtual event, Australia — December 7, 2021
• 📍 NeurIPS MeetUp Cambridge, University of Cambridge, UK — December 8, 2021

Towards unsupervised wearable representations for longitudinal cardio-fitness prediction

• 📍 MobiUK'21, virtual event, UK — July 6, 2021

Learning Generalizable Physiological Representations from Large-scale Wearable Data

• 📍 CHIL'21, virtual event, USA — April 8, 2021
• 📍 NeurIPS'20 ML4MH workshop, virtual event, Canada — December 12, 2020

AI to model Human Behaviour and Health

• 📍 Jesus College Postgraduate Conference, virtual event, UK — March 5, 2021
• 📍 Barclays Eagle Lab, Cambridge, UK — March 12, 2020

Sequence Multi-task Learning to Forecast Mental Wellbeing from Sparse Self-reported Data

• 📍 KDD'19, Anchorage, USA — August 4, 2019

Passive mobile sensing and psychological traits for large scale mood prediction

• 📍 PervasiveHealth'19, Trento, Italy — May 20, 2019

Deep sequence learning for large-scale inference of human behaviour from mobile sensor data

• 📍 MRC Epidemiology Unit, University of Cambridge, UK — March 5, 2019
• 📍 Ocado, Barcelona, Spain & Hatfield, UK (remote) — July 10, 2019

Fast, Visual and Interactive Semi-supervised Dimensionality Reduction

• 📍 Facebook PhD Open House, London, UK — October 25, 2018

Deep Sequence Learning on Mobile Data for Mood Prediction

• 📍 MobiUK'18, Cambridge, UK — September 13, 2018

Class-based Prediction Errors to Detect Hate Speech with Out-of-vocabulary Words

• 📍 ACL'17 Abusive Language workshop, Vancouver, Canada — August 4, 2017

Detecting Irony on Greek Political Tweets: A Text Mining Approach

• 📍 EANN'15, Rhodes, Greece — September 25, 2015

🗞️ Press

---

Audio AI for COVID-19: Cambridge University (1), (2), (3), (4), BBC, The Guardian, Financial Times, The Times, Forbes, Slate, Huffington Post, DailyMail, ITV, IEEE Spectrum, TheNextWeb, STAT, EPFL, TheScientist, The Register, KDnuggets, NPR/WBUR, Royal Society of Biology, Lancet Digital Health, Psychology Today, El Pais, RAI, Corriere della Sera, Focus, DerStandard.

AI for wearables: Cambridge University (1), (2), VentureBeat, Business Insider, Communications of the ACM, National Committee for Quality Assurance.

Big data for psychology: Cambridge University, The Times, CNN, The Telegraph, Sky News, ITV, DailyMail, Inc., CTV, ZDF, ABC.ES, ABC.AU, ELLE, Cosmopolitan, RTBF, TEDx.

Interviews: IndiaAI.gov

🎒 Mentoring

---

I find it incredibly stimulating working with university students. Some recent research projects I supervised:

• Chi Ian Tang — Self-supervised learning in timeseries (papers at IMWUT & NeurIPS-W)
• Benjamin Searle — Predicting compulsive behaviours with wearables (paper at MobileHCI)
• Kevalee Shah — Contrastive learning in activity/clinical timeseries (paper at ML4H)
• Yu (Yvonne) Wu — Weakly-supervised learning for activity timeseries (paper at ML4H)
• Shohreh Deldari — Multimodal self-supervised learning (ongoing)
• Chuen Low — Pure attention models for medical timeseries

I have also been a teaching assistant for the following undergraduate courses:

• Machine Learning & Real-World Data (U. of Cambridge)
• Mobile & Sensor Systems (U. of Cambridge)
• Scientific Computing (U. of Cambridge)
• Numerical Analysis (Aristotle University)

🎠 Playground

---

“The next big thing in technology often starts off looking like a toy”

Quantifying name-dropping

Communitypoprefs.com is a data visualization website, where we present every pop-culture reference over the course of 5 seasons of the TV series Community.

Map out your music taste on Spotify

Visualizing my favourite songs on Spotify with dimensionality reduction and anomaly detection. Data essay published in Cuepoint Magazine, Medium's premier music publication.

Children books and childish language?

Text mining Game of Thrones, Harry Potter, Hunger Games and Lord of the Rings books. Data essay featured in Medium's Editor Picks.

Anonymize kids' faces before posting online

Mobile app with face recognition, age estimation, & emotion recognition to blur kids or replace their face with emotion-based emoji. Developed during HackZurich 2018.

Discover top local news globally

Glocalne.ws was a mashup of Google News and Google Maps. Unfortunately it is now defunct due to API discontinuance.

Composing music and text with Recurrent Neural Networks

Training neural networks on massive amounts of musical notation and literature and letting them create their own art. Essay in Greek but you can still see/listen to the results.

🕳️🐇 Personal

---

Non-academic things about me: I love music, both playing and listening. I am mostly into art rock and indie folk, with the occasional exception of some well-crafted pop. Although I am an accordionist by training, over the last few years I've been playing mostly piano and ukulele. In a previous life, I performed with the critically acclaimed band The Children of the Oldness (aka Kore Ydro) and recorded the album "Consortium in Amato" (listen here).

I also enjoy street photography and in particular playing with light—photography comes from Greek φως (light) and γραφή (writing), or drawing with light. A sample of my shots is on Flickr and one of my landscapes was featured in the Huffington Post.

Lastly, and perhaps most importantly, I'm always on the lookout for ways to move items from the "non-academic list" to the "academic list"—let me know if you'd like to help!