
- This event has passed.

AI Research Talk with Sebastian Gehrmann (Google Research, AI Language)
August 12, 2020 @ 6:00 pm - 7:30 pm EDT
Please RSVP on LinkedIn for this event at: https://www.linkedin.com/events/airesearchtalkwithsebastiangehrmann-googleresearch/
This event will be streamed live so kindly RSVP on LinkedIn to receive the weblink.
We are delighted to convene an AI research talk on Human-AI collaboration for Natural Language Generation with interpretable neural networks. This talk will be presented by Sebastian Gehrmann, Research Scientist in the AI Language group in Google Research.
Title: Human-AI Collaboration for Natural Language Generation with Interpretable Neural Networks
Abstract: Using computers to generate natural language from information (NLG) requires approaches that plan the content and structure of the text and actualize it in fluent and error-free language. The typical approaches to NLG are data-driven, which means that they aim to solve the problem by learning from annotated data. Deep learning, a class of machine learning models based on neural networks, has become the standard data-driven NLG approach. While these approaches lead to increased performance, they replicate undesired biases from the training data and make inexplicable mistakes. As a result, the outputs of deep learning NLG models cannot be trusted. We thus need to develop ways in which humans can provide oversight over model outputs and retain their agency over an otherwise automated writing task. One such approach is to design models as team members instead of autonomous agents, which can be achieved by considering the interaction design as an integral part of the machine learning model development. Sebastian identifies two necessary conditions of team member-models – interpretability and controllability. The models need to follow a reasoning process such that human team members can understand and comprehend it. Then, if humans do not agree with the model, they should be able to change the reasoning process, and the model should adjust its output accordingly. In this talk, Sebastian will demonstrate approaches on how to develop models with these properties.
Schedule:
6:00 PM – 6:10 PM – Welcome and Announcements
6:10 PM – 6:50 PM – Presentation by Sebastian Gehrmann
6:50 PM – 7:30 PM – Q&A with Sebastian Gehrmann
You are cordially invited to join us online for this “must-attend” event.
Profiles of the host and presenter:
• Vik Pant – https://www.linkedin.com/in/vikpant
• Sebastian Gehrmann – https://www.linkedin.com/in/sebastian…
Website:
• Sebastian Gehrmann – https://sebastiangehrmann.com/
Join Synthetic Intelligence Forum online:
• Website – http://www.synthint.ai
• LinkedIn (Page) – http://network.synthint.ai
• LinkedIn (Group) – http://community.synthint.ai
• YouTube – http://video.synthint.ai
Special Thanks to our Partner:
• ET Business Services