Search – talk-data.com

Title & Speakers	Event
PyData Exeter #4: April 17 @ Generator Hub 2024-04-17 · 17:30 Join us at the Generator Hub to enjoy three talks from our brilliant speakers after pizza, drinks, and a chance to get to know the other guests. We then move on to the Topsham Brewery for a chance to socialise a bit more. We look forward to welcoming familiar faces and, hopefully, many new ones. PyData Exeter is popular, so if your plans change, please cancel your attendance on Meetup in good time to free up your space for other community members! Agenda: 18:30 Doors open (ring reception) 19:15 Talks start 20:30 Moving on to Topsham Brewery If you are interested in presenting at future events we'd love to hear from you! Please get in touch via our PyData Exeter talk proposal form. Also, if you have any announcements for the community, please share them. Talks: Modern Data Platforms - Andy Keel Embedding Domain-Specific Knowledge through Reward Shaping in Reinforcement Learning in Football - Vignesh Mohanarajan Python & Ladders: An Introduction to Monte Carlo Modelling - Richard Warburton Modern Data Platforms As we all know, data is valuable. However as organisations have become aware of this, and collected more data, they face the challenge of effectively using it. As data volumes have grown and AI use-cases demand accurate and timely information, new technologies are evolving to enable organisations to use data to it's maximum effect. In this talk we'll look at the concepts of data warehouses, lakes and lakehouses to understand the fundamental data architecture used by leading organisations. Embedding Domain-Specific Knowledge through Reward Shaping in Reinforcement Learning in Football Reinforcement learning is a field of machine learning research that has seen consid- erable growth in the last few years. Reinforcement learning is the process of having an AI-driven system learn through trial and error using feedback from its actions. This feedback is determined by a reward function that is relevant to the environment the system is navigating. However, in certain environments, the feedback can be infrequent leading to a lack of information from which the system can learn. This project proposes a novel reward shaping method by embedding domain-specific information in the form of advanced real-world football metrics in the reward function to solve the aforementioned challenge. This is demonstrated in the Google Research Football environment. Python & Ladders: An Introduction to Monte Carlo Modelling Monte Carlo modelling is a computational technique used in various fields, including finance, engineering, physics, and statistics, to simulate and analyse complex systems or processes. It works by generating random samples or simulations of the system's inputs, often following specified probability distributions, and then approximating the systems behaviour by collating results over numerous samples. With its origins in the Manhattan Project, the method is named after the famous Monte Carlo Casino in Monaco, reflecting the element of chance inherent in the technique. This talk introduces Monte Carlo modelling through the random game of Snakes & Ladders and simple Python code, before exploring real-life applications. Speakers: Andy Keel: Andy works as an Associate Principal at Kubrick Group, where he enables large organisations to leverage their datasets and helps graduates learn the skills necessary for a career in data. Vignesh Mohanarajan: Vignesh is a BSc Computer Science student at the University of Exeter. He is extremely passionate about using data science in sports. Richard Warburton: Richard is a data analyst and data scientist specialising in helping non-technical people understand complex data and analysis. Currently, Richard is contracting with the London Stock Exchange Group and helping them analyse and visualise their contractual data and relationships with suppliers. He also has a sideline in communication training (Vocate) where he trains in public speaking, data literacy, analysis and other essential skills. Sponsors: NumFOCUS, digiLab, University of Exeter CODE OF CONDUCT The PyData Code of Conduct governs this meetup (https://pydata.org/code-of-conduct/). To discuss any issues or concerns relating to the code of conduct or behaviour of anyone at the PyData meetup, please contact the PyData Exeter organisers, or you can submit a report of any potential Code of Conduct violation directly to NumFOCUS (https://numfocus.typeform.com/to/ynjGdT).	PyData Exeter #4: April 17 @ Generator Hub

Join us at the Generator Hub to enjoy three talks from our brilliant speakers after pizza, drinks, and a chance to get to know the other guests. We then move on to the Topsham Brewery for a chance to socialise a bit more. We look forward to welcoming familiar faces and, hopefully, many new ones.

PyData Exeter is popular, so if your plans change, please cancel your attendance on Meetup in good time to free up your space for other community members!

Agenda:

18:30 Doors open (ring reception)
19:15 Talks start
20:30 Moving on to Topsham Brewery

If you are interested in presenting at future events we'd love to hear from you! Please get in touch via our PyData Exeter talk proposal form.

Also, if you have any announcements for the community, please share them.

Talks:

Modern Data Platforms - Andy Keel
Embedding Domain-Specific Knowledge through Reward Shaping in Reinforcement Learning in Football - Vignesh Mohanarajan
Python & Ladders: An Introduction to Monte Carlo Modelling - Richard Warburton

Modern Data Platforms As we all know, data is valuable. However as organisations have become aware of this, and collected more data, they face the challenge of effectively using it. As data volumes have grown and AI use-cases demand accurate and timely information, new technologies are evolving to enable organisations to use data to it's maximum effect. In this talk we'll look at the concepts of data warehouses, lakes and lakehouses to understand the fundamental data architecture used by leading organisations.

Embedding Domain-Specific Knowledge through Reward Shaping in Reinforcement Learning in Football Reinforcement learning is a field of machine learning research that has seen consid- erable growth in the last few years. Reinforcement learning is the process of having an AI-driven system learn through trial and error using feedback from its actions. This feedback is determined by a reward function that is relevant to the environment the system is navigating. However, in certain environments, the feedback can be infrequent leading to a lack of information from which the system can learn. This project proposes a novel reward shaping method by embedding domain-specific information in the form of advanced real-world football metrics in the reward function to solve the aforementioned challenge. This is demonstrated in the Google Research Football environment.

Python & Ladders: An Introduction to Monte Carlo Modelling Monte Carlo modelling is a computational technique used in various fields, including finance, engineering, physics, and statistics, to simulate and analyse complex systems or processes. It works by generating random samples or simulations of the system's inputs, often following specified probability distributions, and then approximating the systems behaviour by collating results over numerous samples. With its origins in the Manhattan Project, the method is named after the famous Monte Carlo Casino in Monaco, reflecting the element of chance inherent in the technique. This talk introduces Monte Carlo modelling through the random game of Snakes & Ladders and simple Python code, before exploring real-life applications.

Speakers: Andy Keel: Andy works as an Associate Principal at Kubrick Group, where he enables large organisations to leverage their datasets and helps graduates learn the skills necessary for a career in data.

Vignesh Mohanarajan: Vignesh is a BSc Computer Science student at the University of Exeter. He is extremely passionate about using data science in sports.

Richard Warburton: Richard is a data analyst and data scientist specialising in helping non-technical people understand complex data and analysis. Currently, Richard is contracting with the London Stock Exchange Group and helping them analyse and visualise their contractual data and relationships with suppliers. He also has a sideline in communication training (Vocate) where he trains in public speaking, data literacy, analysis and other essential skills.

Sponsors: NumFOCUS, digiLab, University of Exeter

CODE OF CONDUCT The PyData Code of Conduct governs this meetup (https://pydata.org/code-of-conduct/). To discuss any issues or concerns relating to the code of conduct or behaviour of anyone at the PyData meetup, please contact the PyData Exeter organisers, or you can submit a report of any potential Code of Conduct violation directly to NumFOCUS (https://numfocus.typeform.com/to/ynjGdT).

Activities & events