Trzy Lipy 3
80-376 Gdańsk
Poland
Miejsce:
GPNT
ul. Trzy Lipy 3, Gdańsk
Building C, Room ABC
We are happy to welcome you to 28th edition of PyData Trójmiasto! We are going to listen to Kreshnaa Raam talking about a proper approach to AutoML and Grzegorz Jacenków talking about Video-Text retrieval methods.
If you have previously signed up to early access to zerve.ai platform for data scientists, then it is a great opportunity to get to know the creators!
When: 21st November, at 18:00 CET
Where: Gdańsk Science-Technology Park, Building C, Room ABC
Registration: The event is free to enter. 100 seats available. Let us know you’re coming via meetup.
Agenda:
18:00 – 18:05 – Meeting boarding
18:05 – 18:10 – A few words about PyData
18:10 – 18:15 – Aga Myśliwczyk will spread a news about her initiatives!
18:15 – 19:00 – AutoML as it should have always been by Kreshnaa Raam
19:00 – 19:05 – A quick break
19:05 – 19:50 – Bridging Text and Video: An Affordable Landscape of Video-Text Retrieval Methods by Grzegorz Jacenków
19:50 – Pizza & Networking
About „AutoML as it should have always been”:
A new modular opensource approach to AutoML and ML pipeline generation in python for different use case/target types. We will go through in-depth understanding of emerging data science platform https://www.zerve.ai/. PyData community received limited access to the platform earlier this year. Now you will have an opportunity to watch the progress and ask all the necessary questions to help us shape the future of data science tooling.
About Kreshnaa Raam:
Kreshnaa is a Lead Data Scientist at Zerve.ai. He works on developing python packages, engineers solutions and conducts general research of the product. Prior to his current experience he worked as McKinsey’s analytics practice contultant and as a data scientist at DataRobot in AI Execution team.
About „Bridging Text and Video: An Affordable Landscape of Video-Text Retrieval Methods”:
Video-Text Retrieval (VTR) has emerged as a pivotal domain in the realm of multimedia analysis and understanding. This presentation delves into the multifaceted applications of VTR, spanning video search engines, content recommendations, video annotations, medical imaging, and beyond. We aim to provide a comprehensive overview of the VTR landscape, encapsulating datasets, models, and benchmarking strategies. Our primary focus revolves around the pivotal issue of affordability, a critical consideration in contemporary VTR research. Recent advancements within the field have predominantly relied on resource-intensive large-scale models, demanding extensive GPU power and extensive datasets. By scrutinising the latest innovations and their economic implications, we aim to shed light on the balance between computational demands and real-world applicability in VTR.
About Grzegorz Jacenków:
Currently a Data Scientist at Ring, Grzegorz Jacenków specialises in multimodal learning research and large language models (LLMs). Prior to joining Amazon, he was a PhD student in Healthcare AI at The University of Edinburgh, where he also earned an MSc in Artificial Intelligence. His academic foundation was laid with a BSc in Computer Science with Business and Management from The University of Manchester. Notably, Grzegorz contributed to CERN as a technical student, addressing author disambiguation at Inspire-HEP. His research interests encompass multimodal alignment, low-resource learning, and leveraging knowledge graphs.
Notes:
The event will not be live-streamed nor recorded.
https://www.meetup.com/PyData-Trojmiasto/events/297201599