Federated Learning for All: Crowdsourced Machine Learning on its Journey to Privacy & Fairness
Wednesday, September 16, 2020 (Central Time) — 12:30PM - 1:45PM
Personal and device data is growing at a mind-boggling rate, and effectively utilizing such data is of increasing interest to the machine learning (ML) community. Meanwhile, there is greater focus on reducing the centralized storage of potentially sensitive data, maintaining user privacy, and avoiding magnifying biases present in data in trained models. In this workshop, we will discuss Federated Learning (FL), an ML setting where a network of devices collaboratively train a central model on-device without sending their raw user data to a central server. Attendees should come away with an interest to learn more about FL, as well as an understanding of (i) basics of ML and FL, (ii) some of the challenges inherent in this ML setting, and (iii) the intersection of privacy and fairness in FL. Time permitting, we will pivot slightly In the last portion of the session and focus on the speaker’s journey, Jenny Hamer, and discuss why it is important to pursue fairness while considering representation in the industry. We will host a brief panel and highlight her particular experiences and career path as a professional, and explore how to create an environment that fosters diversity and inclusivity; with Google AI Residency as an example program that aspires to design for inclusivity in AI and research.
Jenny Hamer, Nina Ong