PyData Sydney: Inside ML Models with SHAP - Ansh Bordia

PyData Sydney: Inside ML Models with SHAP - Ansh Bordia
Timestamps:
00:00 Stream starting
01:00 Welcome
01:24 Intro Slides
05:45 Presentation Agenda
06:33 The Need For Explainable AI
08:20 What is SHAP?
09:27 Shapley Values Intuition
13:00 Problems with SHAP?
16:09 Missing Features
17:42 Question Time
19:54 SHAP Python Library
21:39 Practical Example
23:05 Feature Effects Plot
24:10 Partial Dependence Plot
25:00 Local Explainability
26:43 Advanced uses of SHAP
28:42 Question Time
36:00 Closing
Machine learning models are often described as black boxes. This leaves developers unsure of why they work the way they do.
With SHAP we go inside the black box to understand how the ML models make predictions.
In our meet up we will discuss:
- What is SHAP?
- SHAP and the link to Game Theory
- Using SHAP in a business setting
- Advanced uses of SHAP
About the Speaker:
Ansh is passionate about making useful products using Machine Learning and open source contribution. He also loves reading science magazines, especially articles on astrophysics and climate change. Ansh authors data science articles as well to help others.
--------
Online Events:
--------
We will be using RUclips streaming to host this event. That way you can interact through live chat to ask questions. Also the recording will be made available shortly after the event has finished if you are unable to attend.
--------
Connect with PyData Sydney Community:
--------
Twitter:
/ pydatasydney
LinkedIn:
/ pydata-sydney
Meetup:
www.meetup.com...
--------
About PyData:
--------
PyData is an educational program of NumFOCUS, a 501(c)3 non-profit organization in the United States. PyData provides a forum for the international community of users and developers of data analysis tools to share ideas and learn from each other. The global PyData network promotes discussion of best practices, new approaches, and emerging technologies for data management, processing, analytics, and visualization. PyData communities approach data science using many languages, including (but not limited to) Python, Julia, and R.
NumFOCUS Code of Conduct : numfocus.org/c...
- PyData is dedicated to providing a harassment-free meeting experience for everyone, regardless of gender, sexual orientation, gender identity and expression, disability, physical appearance, body size, race, or religion.
- We do not tolerate harassment of meeting participants in any form.
- All communication should be appropriate for a professional audience including people of many different backgrounds.
- Sexual language and imagery is not appropriate for any conference venue, including talks.
- Be kind to others.
- Do not insult or put down other attendees.
- Behave professionally.
- Remember that harassment and sexist, racist, or exclusionary jokes are not appropriate for PyData.