On June 20th, 2023, 80 trainees who participated in the AI4Good Lab initiative, which aims to make AI education more inclusive, presented at Mila the 15 machine learning prototypes they built over three weeks, applying knowledge gained from the program’s curriculum and exploring how AI can be used to address societal challenges.
Each team presented their project to over 150 attendees during Demo Day, the culmination of 7 weeks of dedicated work bringing together numerous teaching assistants, lecturers, speakers, mentors, sponsors, and experts who supported the year’s cohort of future leaders as they encountered technical and ethical challenges along the way.
Beyond machine learning lectures, trainees also attended talks and workshops designed to showcase real-life applications of AI and put their knowledge into practice.
Topics included responsible AI, privacy, entrepreneurship, and trainees attended a panel with the team developing Biasly, a project that started at the AI4Good Lab in 2018 that is now run at Mila. They also piloted the Indigenous AI module developed on behalf of CIFAR, learned about Mila’s First Languages AI Reality (FLAIR) project, and took part in practical workshops designed to guide them through the critical steps of building a machine learning prototype as a team.
From health to environment to Indigenous languages, trainees identified global issues and worked in teams to build solutions to address them. They designed AI systems to help with reducing food waste, recycling and disposing of garbage more effectively, to suggest which plants would be more suitable to certain urban environments or to help make more responsible AI design decisions. They also built apps to foster Indigenous language communications, help tackle biased and problematic online content, and analyze ultrasound in real time to better care for expectant mothers.
A wealth of perspectives
The team behind the BiodiversCity project shared reflections on their time at the Lab: “Being a part of the AI4Good Lab has been a truly amazing experience. Each of us came into the program with our own set of worries: how to make friends, whether our backgrounds were sufficient, and whether it would even be possible to create a real machine learning project. What a pleasant surprise to arrive and find that every person in the Lab was friendly and working hard to make it fun”
Dhwani Patel, who participated in the AF Discovery project in Toronto, said: “AI4GoodLab’s experience has been a game-changer in my professional journey. From insightful workshops on AI/ML to invaluable networking opportunities and the hands-on project experience has opened a wide range of career opportunities. It has not only provided a strong foundation but also fueled my growth, inspiring me to reach new heights in the field of AI. AI4GoodLab is the perfect place to kickstart your career and embark on a transformative journey of learning, development, and endless possibilities.”
Christina Isaicu, AI4Good Lab Associate, Operations and Research at Mila, said that the whole point of the event is for trainees to feel like they belong.
“The people who are affected by these tools need to have a hand in shaping them,” she said.
“There is a pressing need to have a wealth of perspectives in the field of AI, from disciplinary backgrounds to gender, race, sexuality, ability, class, and a multitude more of experiences and their intersections,” she added.
Converging to Montreal
This was the first year that Mila – Quebec AI Institute was running operations and the Montréal program following AI4Good Lab’s integration in January 2023. The program was delivered online and in person in 3 Canadian cities: Montréal at Mila, Edmonton, in partnership with Amii and Toronto, in partnership with CIFAR and Vector Institute with in-kind contribution from the Design Fabrication Zone (DFZ) at Toronto Metropolitan University.
This year’s event was a turning point for the AI4Good Lab: the return to in-person activities through a fully hybrid program delivery allowed trainees to foster meaningful connections and physically collaborate again. It was represented across 7 provinces and 29 universities as part of its commitment to the pan-Canadian AI strategy to increase training in AI across the country.
The Lab’s founding partners, CIFAR and OSMO, provide the funding for the program that enables the Lab to continue growing this supportive community.AI4GoodLab also thanks its delivery partners Amii, Vector Institute and Design Fabrication Zone, our title sponsors DeepMind and IVADO, and our Key Sponsor, Manuvie.
The Accelerator Award
One team from each cohort was selected for the Accelerator Award, an opportunity to extend the project with continued support from the Lab network.
Montreal Accelerator Award Recipients
Make better recycling decisions with Recyclo – our AI-powered image detection guides your waste to the right destination.
Team members: Blue Namba, Rency Waneskehian, Zhen Xu, Dejun Yang, Haiqi Zhou
Toronto Accelerator Award Recipients
SafeT.First harnesses the power of ML models and historical crime data to provide Toronto’s pedestrians with safer route options, ensuring peace of mind on every walk.
Team members: Chloe Abbruzzese, Yllka Bojku, Cynthia Lam, Harjot Kaur Grewal, Samira Fairuz Ahmed, Dhwani Patel
Edmonton Accelerator Award Recipients
Say goodbye to guesswork: PlatePal analyzes all available choices to create a personalized and efficient grocery plan.
Team members: Urim Iyasere, Sana Shams, Aesha Patel,Tanya, Anh Vo