[email protected] conferences: Math for Data Science ran on December 1, 2020.
[email protected] conferences are one-day technology conferences from Manning Publications. Manning authors and other industry experts deliver live coding sessions, deep dives, and tutorials. Watch for free on Twitch.
talks from experts
Featuring expert speakers, plus ten minute lightning talks.
0:00 – 28:28 The complex inner life of simple regression | Matthew Rudd, author of “Python Concurrency with asyncio”
28:29 – 38:02 Living in another Dimension – Principal Component Analysis (PCA) | Nicole Koenigstein
38:03 – 1:05:52 Beyond MCMC: Drawing Samples is Only One Way to Do Probabilistic Modeling | Brian Godsey, author of “Think Like a Data Scientist” and “Exploring the Data Jungle”
1:05:53 – 1:33:47 What really is data science? | Anh Le
1:33:47 – 1:44:04 Linear and non-linear models for humanoid posture control and balance | Vittorio Lippi
1:44:05 – 2:21:49 Bayesian non-parametrics, bootstrapping and the Chinese Restaurant Process | Alexandra Posekany
2:21:50 – 2:30:57 What are the methods to calculate “the middle value” of points and when to use which | Ines Dedović, co-author of “Algorithms and Data Structures for Massive Datasets”
2:30:58 – 2:51:55 How do we get computers to draw faces? | Luis Serrano, author of “Grokking Machine Learning”
Missed [email protected] conferences: Math for Data Science? Rewatch now on Manning’s YouTube channel.
conference speakers
Brian Godsey
Brian Godsey is a mathematician and data scientist working with start-ups and high-growth companies to build highly statistical software products that provide actionable insight to end users, often in areas with never-before-seen data or with no existing solutions at all.
Probabilistic modeling is a set inference techniques that treats uncertainty as an integral part of a system rather than something to avoid. Bayesian statistics lays the foundations and advanced optimization gives us the tools to build models, but how much do these models really tell us about the underlying system and the uncertainties involved? This talk starts from a simple example of a non-probabilistic model and sequentially adds elements of statistical uncertainty, focusing on what we gain or lose with each addition.
Nicole Koenigstein
Nicole Koenigstein is a Software Engineer at Refline AG.
Big Data is a buzzword nowadays and with lots of data comes the so called curse of dimensionality. This is where PCA comes in – it is one of the most popular linear dimension reduction methods. Learn why PCA is so powerful and how to implement it in Python.
Matthew Rudd
Matthew Rudd is a mathematician fascinated by statistical modeling, data analysis, and the tensions between theory, practice, and interpretability in data science. He teaches mathematics and statistics at Sewanee: The University of the South, a liberal arts college in Tennessee.
Data science seems obsessed with complicated algorithms and Big Data these days, but simple models often provide the most useful insight, especially if people need to interpret the results. Even simple models can be surprisingly sophisticated, though, so it’s essential to understand the underlying assumptions and their implications. I’ll discuss some common misconceptions about regression and how to combat them with mathematics.
Alexandra Posekany
Alexandra Posekany is a University Assistant (PostDoc) at The University of Technology Vienna
The notions of Bayesian statistics have become more prominent in machine learning. We will look specifically at the mathematics behind Bayesian non-parametrics, starting out with the Bayesian Bootstrap and moving to Dirichlet processes.
Vittorio Lippi
Vittorio Lippi is a roboticist working on benchmarking for exoskeletons and humanoids, and human posture control. He has published 40 peer reviewed works on journals and conference proceedings.
Human posture control models are used to analyse neurological experiments and control of humanoid robots. This presentation focuses on the nonlinear posture control model, the DEC (Disturbance estimate and Compensation) used in order to compensate disturbances, based on signals coming from sensor fusion. The implications of the non-linear nature of the DEC are discussed in terms of system dynamics and system identification.
Luis Serrano
Author of Grokking Machine Learning, and Quantum AI Research Scientist
at Zapata Computing
Have you ever wondered how computers draw those very realistic looking faces? This is a branch of machine learning called generative machine learning.
In this talk we’ll give an overview of what generative machine learning is, what the main algorithms are, and in particular, we’ll delve into one of the most popular ones: Generative adversarial networks (GANs). We’ll build a very simple pair of GANs that will generate a small dataset of images. No previous knowledge of machine learning or neural networks is required, as we’ll introduce everything we use in the talk.
Ines Dedovic
Author of Algorithms and Data Structures for Massive Datasets, and Development Engineer at Jonas & Redmann
What are the methods to calculate “”the middle value”” of points and when to use the quadratic/ arithmetic/ geometric/ harmonic mean, the median, and function fitting and using known integrals
Dr Meltem Ballan
Professional Fellow, Data Science at General Motors
Dr Ballan’s talk will cover Food mould and ground penetrating radar data object recognition. The goal is to summarize the art of the possible in 20 min. Dr Ballan will present the steps of object recognition and classification.
Anh Le
Anh Le is a Data Science Associate, Strategic Foresight at TMX Group. She studied Biomedical Engineering and Finance at the University of Waterloo, and holds a Master in Management Science.
Modern data science emerged in tech, from optimizing Google search rankings and LinkedIn recommendations to influencing the headlines Buzzfeed editors run. But it’s poised to transform all sectors, from retail, telecommunications, and agriculture to health, trucking, and the penal system. Yet the terms “data science” and “data scientist” aren’t always easily understood, and are used to describe a wide range of data-related work.
sponsored by EventBrowse
EventBrowse.com is a comprehensive directory of the best business events and conferences in major metropolitan areas in the US, and a growing number of cities around the globe These events are curated into weekly newsletters, with recommendations for more than 75,000 active subscribers.
