Aristides Milios

MSc, McGill University/MILA

aristides.milios [AT] mila.quebec

Bio

My name is Aristides Milios, and I'm a highly motivated machine learning researcher passionate about the intersection between Natural Language Processing (NLP) and Machine Learning, specifically about investigating the reasoning and few-shot capabilities of Foundation Models (e.g. GPT-3, OPT, etc.), especially when paired with smaller retrieval-style models. I am also passionate about promoting entrepreneurship and design thinking education, at both the secondary and tertiary level.

In high school, I co-founded a startup called Bitness.io, which went through the Launch36 and Propel ICT accelerator programs in Atlantic Canada, and used WiFi beacons to track customers through brick-and-mortar stores and provide Google-Analytics-style analytics to store owners.

As part of my efforts to promote entrepreneurship and design thinking education, I jointly co-founded Hoist Halifax, a club with monthly workshops for teens who are interested in technology and entrepreneurship. I was involved in promoting entrepreneurship and design thinking at Dalhousie University for the duration of my studies, going through the rigorous University Innovation Fellows program, organized by Stanford's d.school of Design, and becoming a Fellow of the program at Dalhousie. As co-president of the Dalhousie Entrepreneurship Society, I organized a series of design thinking workshops on campus, open to students of all faculties.

As part of my undergrad, I did two research internships at MarineTraffic Inc. (summer 2018 and winter 2019), in Athens, Greece, researching the fusion of AIS ship tracking data and radar-based satellite imagery, in cooperation with the Institute for Big Data Analytics at Dalhousie University.

In the fall of 2019, I completed an internship at the Know-Center research institute in Graz, Austria, doing research into Active Learning and an interactive feedback-driven network traffic dataset labelling tool. I completed my Bachelor of Computer Science with First Class Honours at Dalhousie in August of 2020. For a year I did a combination of TA work and research work at Dalhousie under professors from my undergrad. In September of 2021, I began an M.Sc. degree in Computer Science at McGill University/MILA in Montreal, under Dr. Siva Reddy. As of Sept. 2022, I'm also being co-supervised by Dr. Dzmitry Bahdanau.

Publications

Most recent publications on Google Scholar.
indicates equal contribution.

LLMs can learn self-restraint through iterative self-reflection

Alexandre Piché, Aristides Milios, Dzmitry Bahdanau, Chris Pal

ICLR 2024 Workshop on Secure and Trustworthy Large Language Models

In-Context Learning for Text Classification with Many Labels

Aristides Milios, Siva Reddy, Dzmitry Bahdanau

GenBench Workshop at EMNLP 2023

An Analysis of Social Biases Present in BERT Variants Across Multiple Languages

Aristides Milios, Parishad BehnamGhader (equal contribution)

Trustworthy and Socially Responsible Machine Learning (TSRML) Workshop at NeurIPS 2022

Survey of Generative Methods for Social Media Analysis

Stan Matwin, Aristides Milios, Pawel Pralat, Amilcar Soares, Francois Theberge

LLMs can learn self-restraint through iterative self-reflection

Alexandre Piché, Aristides Milios, Dzmitry Bahdanau, Chris Pal

ICLR 2024 Workshop on Secure and Trustworthy Large Language Models

In-Context Learning for Text Classification with Many Labels

Aristides Milios, Siva Reddy, Dzmitry Bahdanau

GenBench Workshop at EMNLP 2023

An Analysis of Social Biases Present in BERT Variants Across Multiple Languages

Aristides Milios, Parishad BehnamGhader (equal contribution)

Trustworthy and Socially Responsible Machine Learning (TSRML) Workshop at NeurIPS 2022

Survey of Generative Methods for Social Media Analysis

Stan Matwin, Aristides Milios, Pawel Pralat, Amilcar Soares, Francois Theberge

Monitoring Marine Protected Areas using Data Fusion and AI Techniques,

Konstantina Bereta, Aristides Milios, Konstantinos Chatzikokolakis, Dimitris Zissis

NATO CMRE Maritime Situational Awareness Workshop 2019

Automatic Fusion of Satellite Imagery and AIS data for Vessel Detection

Aristides Milios, Konstantina Bereta, Konstantinos Chatzikokolakis, Dimitris Zissis, Stan Matwin

2019 22th International Conference on Information Fusion (FUSION).

Timeline

Full resume linked here.

The excellent template for this website was created by Martin Saveski. Click on his name to learn more.