#22 MLMilan: Real Reinforcement learning tasks & managing data science projects

Dear all,

This time we are delighted to present to you with new speakers before the summer vacation.
ML Milan is back at the end of July and we will explore Reinforcement learning applications and management of data science projects.
The event will be held online, in English with our classical format via Zoom platform.

Talk 1: “Making robots learn to perform real-world tasks through Reinforcement Learning”

Reinforcement learning techniques have been applied in a variety of contexts, ranging from playing games to natural language processing. However, RL methods have mainly been limited to non-physical domains. Controlling real world physical systems like robots through RL adds an extra complexity. In this talk, we will see how to make a real-world industrial robot learn to perform a task by using reinforcement learning.

Asad Ali Shahid,
Asad is currently working on a research project with Swiss AI Lab (IDSIA). He is a masters student in mechanical engineering at Politecnico di Milano. His research interests lie at the intersection of robotics and machine learning. In particular, he is interested in collaborative robots that work in close proximity with humans. Before, he studied
robotics at ETH Zurich.

Talk 2: “Data Workspaces: Python-based Management of your Data Science Projects”
Modern scientific workflows can be very complex, involving many data sources, software components, and partial results. At the same time, many workflows are not automated and incur significant manual effort or depend on brittle, one-time, scripts. As a result, scientists and data professionals have issues with managing experiments, collaboration, and reproducibility.

Data Workspaces (DWS) is a Python-based open source framework for managing scientific data and automating experiment workflows. Data Workspaces maintains the state of a (data) science project, including data sets, intermediate data, results, and software. It supports reproducibility through snapshotting and lineage tracking and collaboration through a push/pull model layered on top of the Git version control system.

In this talk, we will give an overview of Data Workspaces, show a live demo of Data Workspaces running in a Jupyter Notebook, and describe how some projects have been organized and tracked in DWS.

Jeff Fischer
Jeff is CTO of Benedat LLC, a Data Science company in Silicon Valley focused on data intensive systems, from infrastructure to machine learning. He advises data engineering and data science teams on architecture and technology selection. Jeff has a PhD in Computer Science from UCLA.

Rupak Majumdar
Rupak is a Scientific Director at the Max Planck Institute for Software Systems in Kaiserslautern, Germany. His research interests are in the verification and control of reactive, real-time, hybrid, and probabilistic systems, software verification and programming languages, logic, and automata theory. He received the B.Tech. degree in Computer Science from the Indian Institute of Technology at Kanpur and the Ph.D. degree in Computer Science from the University of California at Berkeley.

Where? Join us on Thursday 30th of July in our online event via Zoom, at this link. The event will start at 18:45 and will finish around 20.00

Link Zoom Meeting:
https://us02web.zoom.us/j/83287927486?pwd=eE5QVFdqeklnbngwa00zcm1wZFc1Zz09

Meeting ID: 832 8792 7486
Password: 343766