By Jonathan Woods
July 21, 2022
The inaugural cohort of Mothers & Machine Learning (M&ML) ― a professional development program tailored to serve mothers on maternity leave and stay-at home caregivers ― featured an incredible group of students from across professional and educational backgrounds. Despite their different accomplishments, they all shared two things in common: a strong drive to learn about AI and first-hand experience with the challenges of building professional skills while shouldering childcare responsibilities.
Over six weeks, students heard live lectures from instructors and guest speakers, built familiarity with core machine learning concepts, and applied their learnings to real-world problems of their choosing. No previous machine learning experience was required.
Most importantly, the program was designed with the specific needs of mothers and caregivers in mind. This involved providing:
- Personalized support from TAs: In addition to a main instructor, four teacher’s assistants (TAs) were available to answer questions, go over assignments, and review solutions. In a program of 45 students, this high instructor-to-student ratio allowed for highly personalized support. Course Manager Melissa Valdez praised the dedication and quality of the TAs, three of whom were Ph.D. students while the other was an innovation-focused project manager. “The TAs were super committed to this course,” said Valdez., “We were very lucky to have four incredible TAs ― Flora Wan, Matthew Duschenes, Mohamed Hibat Allah, and Schuyler Moss.”
“The TAs were extremely knowledgeable and patient. Truly, the people involved led to the success of the program. I’ve never felt so supported in a course.” Ingrid Giesinger, PhD Student, Health Industry
- Scheduling flexibility: While the online classes were conducted live to foster interaction, recordings were always quickly made available for anyone with competing time demands. Because the team understood the flexibility that mothers and caregivers require, they also opened two office hour periods for one-on-one time. “It was all about meeting students where they are,” Valdez said.
- Eligibility for $500 in childcare funding: Students that completed the program were eligible to receive $500 in childcare funding, made possible through the Google TensorFlow Faculty Award. Awarded to course creators Juan Felipe Carrasquilla Álvarez (also lead instructor) and Dora Astrid Gaviria Castaño, this grant is provided to support machine learning education programs for underrepresented groups.
- A close and supportive community: Students connected and networked through a Slack channel and a study room where they could self-organize and work together on assignments. A regular Tuesday “Coffee and Chat” event sprung up organically, becoming a weekly meetup in which students shared professional and personal experiences. From AI assignments and career insights to childcare and post-partum experiences, all important topics for this group were open for discussion and met with support, stories, and humour.
“I loved learning alongside other moms who ‘get it.’” Grace Yan, Manager in Government
“It was so inspiring to meet other mothers.” Si-Yau Que, Senior Supply Chain Analyst
M&ML was created to address two priorities for Vector. The first was to enable those who often unfairly experience a ‘motherhood penalty’ to participate in the fast-growing field of AI. Challenges that mothers can face include a lack of sufficient support during maternity leave and, after returning to work, a lack of reintegration plans to help catch up with changes or new technologies, a reduction in responsibilities, and bias that can lead to derogatory comments. In contrast, M&ML made sure the needs of motherhood came first. The second priority was to add to Ontario’s AI talent pool amid the fierce global competition to attract and retain AI practitioners.
Over the course of the program, M&ML students studied conceptual AI-related questions, optional math proofs, and Python coding problems with a focus on computer vision and natural language processing applications. Learning culminated in a capstone project, which involved the application of course material to a problem of each student’s choosing. Final projects from the inaugural cohort included work on:
- Retail sales: using grocery purchasing data from Instacart to predict future orders
- Health care: predicting the probability of heart attack given a set of personal and medical attributes
- Marketing: testing different machine learning models to improve conversion rates in directed marketing campaigns
Despite the program’s rigour, instructors strove to imbue the experience with a spirit of growth and empathy. The motto, “If you don’t know how something works, that’s an opportunity,” ― a favourite of Carrasquilla Álvarez’s ― encapsulated the program’s culture.
“We just created an atmosphere that got rid of all those fears of asking questions and not knowing basic stuff,” Carrasquilla Álvarez said. “We also had an amazing TA that would answer questions during the lectures, so everyone felt supported, no matter how simple the questions were.”
That atmosphere enabled participants ― including those lacking a computer-science background ― to build their confidence working with machine learning. Barra K Al-Khazarji, an Assistant Professor in McMaster’s Department of Kinesiology, said, “This was…a life-changing course for me (and likely many other mothers). This was the exact confidence boost and filling of a knowledge gap that I needed to be able to explore these concepts on my own after the class.”
“This course has stoked a fire in me,” she added.
For many, the program has also opened a door to further AI-related professional development pursuits. The majority of students signed up for a subsequent Excel to AI workshop, which teaches Python programming skills to people who are proficient with Microsoft Excel. Several students have also made profiles on Vector’s Digital Talent Hub, an online job platform that matches job and internship seekers with AI-specific opportunities.
On the expanded professional horizons that M&ML inspires, Susmita Saha, a Sales Advisor who attended the program, said, “This course has given me so much more confidence to pursue my career in the AI field.”
Finally, the most succinct reflection on the support, education, and potential provided by M&ML came from Donna Vakalis, an NSERC Postdoctoral Fellow focused on green buildings engineering, who summed up her experience this way:
“My present-day self wholeheartedly recommends taking this course to my past self.”
Sign up for Vector’s newsletter today to learn when applications open for the Winter 2023 Mothers & Machine Learning course!