Abstract: The recent progress made in Deep Learning have allowed the development of amazing new applications, ranging from self-driving cars to chatbots. However, while those new AI systems excel at one skill, they are still not capable of measuring up to humans. That’s because they lack one critical skill: adaptability. Meta-learning, or learning-to-learn, opens the way to building systems able to continuously acquire new skills to tackle a wide variety of problems, and just like humans, learn new skills by recycling their own knowledge. From LSTM to one-shot learning, this talk attempts to give an overview of the current techniques gathered under the name of ‘meta-learning’, and examines how new research might impact the Data Science field.
Bio: Dr. Jennifer Prendki is the Head of Data Science at Atlassian, where she leads all Search and Machine Learning initiatives and is in charge of leveraging the massive amount of data collected by the company to load the suite of Atlassian products with smart features. She received her PhD in Particle Physics from University UPMC - La Sorbonne in 2009 and has since that worked as a data scientist for many different industries. Prior to joining Atlassian, Jennifer was a Senior Data Science Manager in the Search team of Walmart eCommerce. She enjoys addressing both technical and non-technical audiences at conferences and sharing her knowledge and experience with aspiring data scientists.