Name: ML/AI Conversation: GraphNN and Reasoning Improvement for LLM
Start: 2024-02-06T17:30:00-05:00
End: 2024-02-06T21:00:00-05:00
Location: 11 W 19th St

Dear All,

We are delighted to announce that our upcoming meetup is scheduled for Tuesday, February 6th. We invite you to join us at the Capital One Flat Iron office to discuss the latest advancements in Machine Learning and Artificial Intelligence. There will also be an opportunity to network and enjoy some pizza.

Agenda:
5:30 Doors open,
5:30 - 6:30 reception, networking
6:30 - **Max Eremeev, AI Research Engineer (Elemental Cognition)** \- "Curriculum training to improve reasoning of LLMs"\.
7:10 - **Anton Tsitsulin, Research Scientist (Google)** \- "Large Scale Machine Learning on Graphs"\.
7:50 - 9:00 Further Networking

!!! IMPORTANT !!!
Please provide your full legal name at the time of RSVP, as it is necessary to enter the building and take your photo ID with you!

Anna Leonenko

Vlad Eremin

Koen

New York AI/ML Conversations

Technology

Science

Software Engineering

Engineering

Machine Learning

Data Science

Big Data

AI/ML

Please provide your full legal name for building security

Max Eremeev, AI Research Engineer, Elemental Cognition
​Curriculum training to improve reasoning of LLMs
​At Elemental Cognition we’re trying to tackle complex issues for high-stakes domains, e.g., pharma or investment management. And while everyone is fooled by the fluency of LLMs, they are not applicable a high risk tasks because of hallucinations or degenerate outputs. At EC, we tackle this problem by employing curriculum training — we train hundreds of adapters on top of 7B models, progressively increasing the complexity of use cases. This talk is dedicated to learn more about good practices of curriculum learning and useful training/generation regularization techniques for resolving typical degeneracies.

​Anton Tsitsulin (Google Research)
Large Scale Machine Learning on Graphs
​Graphs are general data structures that can represent information from a variety of domains (social, biomedical, online transactions, and many more). Graph Neural Networks (GNNs) are an exciting way to use graph structured data inside neural network models that have recently exploded in popularity. However, implementing GNNs and running GNNs on large (and complex) datasets still raises a number of challenges for machine learning platforms. The talk will be based on TF-GNN, a library for working with graph structured data in TensorFlow.

​Anton Tsitsulin and Max Eremeev

ML/AI Conversation: GraphNN and Reasoning Improvement for LLM

Artificial Intelligence

Technology Startups

11 W 19th St

COVID-19 safety measures

The event host is instituting the above safety measures for this event. Meetup is not responsible for ensuring, and will not independently verify, that these precautions are followed.

ML/AI Conversation: GraphNN and Reasoning Improvement for LLM

Details

COVID-19 safety measures