Overcoming the Cold Start Problem: How to Make New Tasks Tractable
Toronto Machine Learning Series (TMLS) via YouTube
Google, IBM & Microsoft Certificates — All in One Plan
Power BI Fundamentals - Create visualizations and dashboards from scratch
Overview
AI, Data Science & Cloud Certificates from Google, IBM & Meta — 40% Off
One plan covers every Professional Certificate on Coursera. 40% off your first 3 months — limited time.
Unlock All Certificates
Explore strategies for addressing the cold start problem in machine learning during this 45-minute Toronto Machine Learning Series (TMLS) conference talk. Delve into the challenges faced by AI companies when scaling their operations, particularly the reliance on large, high-quality datasets for training deep neural networks. Learn how to mitigate the time and cost constraints associated with collecting extensive labeled data by examining techniques for aggregating information across multiple sources and leveraging pre-trained models. Gain insights from speakers Azin Asgarian, Applied Research Scientist at Georgian, and Franziska Kirschner, Research Lead at Tractable, as they demonstrate practical approaches to make new tasks more tractable and accelerate the time to value for businesses implementing AI solutions.
Syllabus
Overcoming the Cold Start Problem: how to Make New Tasks Tractable
Taught by
Toronto Machine Learning Series (TMLS)