Pie & AI Montevideo: Machine Learning Interpretability & Learning to Rank

El 27 de julio se llevará a cabo un nuevo evento de Pie & AI de la comunidad de deeplearning.ai en Montevideo. El evento será presencial.

Este evento forma parte de la serie Pie & AI de deeplearning.ai: https://www.deeplearning.ai/events/

El evento tiene cupos limitados. Para participar, deben inscribirse a través de la lista de attendees de Meetup.

En caso de completarse los cupos quedará habilitada una lista de espera. Los invitamos a anotarse y les confirmaremos en caso de liberarse un espacio.

👉 Contenidos

• Machine Learning Interpretability

• Speaker: Ignacio Aristimuño.
• Resumen: Data has become the new electricity, but with great power there must also come great responsibility. Hidden biases in data, social acceptance of Machine Learning models and having a better understanding of how they work and make predictions, are just a few of the main reasons that AI interpretability and explainability are becoming a trend. In this talk we will explore what interpretability is, why it truly matters and which techniques we can use to enhance the social acceptance of Machine Learning models. It will also cover the detection of bias-related issues in data.

• Learning to Rank

• Speaker: Michel Helal.
• Resumen: Ranking is everywhere. Our daily news are ranked by some algorithm before reaching our smartphone. Our search results are ranked by each search engine on the web (not always with the same accuracy, sadly) . This is the era of big data, a seemingly unlimited flux of information knocks on our door every day. But our time is limited, and also our brain capacity. We need help to figure out what information is important, and what is meaningless. Learning to rank (LTR) leverages machine learning power to help search engines in the task of ranking. In this talk we will approach ranking as an optimization problem. From then, it is straightforward to end talking about ML. We will discuss the details of this powerful technique, and show how it is capable of capturing users behavior, transforming the ranking accordingly. To make things more concrete we will describe LTR ideas in the context of our current work at PedidosYa, where we are aiming to enhance our restaurant's shop list using this technique.

👉 Sobre los oradores:

Ignacio Aristimuño: Production Engineer with a background in logistics and mathematical optimization, currently pursuing a Master’s in Big Data. Passionate for AI and its applications for generating a positive impact in our lives.

Michel Helal: Once a scientist, today a data scientist. I’ve spent almost 7 years enjoying one of my child’s dreams: working at a physics department in a research team. The other dream, inspired by Isaac Assimov (nothing original here) became true some years ago when I joined the AI department at Antel as a Data Scientist. Now this dream continues growing at PedidosYa, under the same role. I have a bachelor's degree in physics from Udelar and I am currently enrolled in a MSc degree in artificial intelligence from UNIR, Spain. Huge fan of Science in general, and of data science in particular. I also think “AI is the new electricity” and I am honored and truly excited to be part of this revolution.

👉 Sponsors:

Marvik, Moove It y deeplearning.ai 🤖

¡Los esperamos!

This is a Pie & AI event in partnership with DeepLearning.AI. Pie & AI is a series of DeepLearning.AI meetups independently hosted by community groups. This event is hosted by Marvik. In addition to signing up on Meetup.com, sign up here (https://www.eventbrite.com/e/157288218295) if you'd like to receive a course promo code and future event updates from DeepLearning.AI