Date: |
02 Dec 2024, 11.00AM – 12.00PM |
duration: |
1 hr |
Location: |
Online |
Cost: |
Free event |
Prompt engineering refers to the understanding and techniques that underpin effective prompting of a Generative AI. A wide taxonomy of prompting techniques has rapidly emerged and continues to be an area of active research. However, for most use cases, there are three techniques that are likely to be most helpful. And an understanding how these techniques work can tell you a lot about the underlying language model.
This webinar will introduce you to the mechanics and rationale of ABS (always-be-specific) prompting, few-shot prompting, and chain-of-thought prompting. Along the way, we’ll learn about semantic meaning, next-token prediction, and scale-driven emergence in AI systems. There will be lots of example prompts to try, so make sure you have a browser opened to your AI of choice!
David Dempsey is an Associate Professor at the University of Canterbury’s Department of Civil and Natural Resources Engineering. He delivers Generative AI training for the 800 students within the department, as well as staff across the wider campus. His research group uses Artificial Intelligence to provide advanced warning of volcanic eruptions, floods and wildfires, while another part of the group works on computational engineering solutions for subsurface energy systems: geothermal, carbon removal, and hydrogen storage. He has previously worked at Los Alamos National Laboratory, Stanford University, and the University of Auckland.
Prompt engineering refers to the understanding and techniques that underpin effective prompting of a Generative AI. A wide taxonomy of prompting techniques has rapidly emerged and continues to be an area of active research. However, for most use cases, there are three techniques that are likely to be most helpful. And an understanding how these techniques work can tell you a lot about the underlying language model.
This webinar will introduce you to the mechanics and rationale of ABS (always-be-specific) prompting, few-shot prompting, and chain-of-thought prompting. Along the way, we’ll learn about semantic meaning, next-token prediction, and scale-driven emergence in AI systems. There will be lots of example prompts to try, so make sure you have a browser opened to your AI of choice!
David Dempsey is an Associate Professor at the University of Canterbury’s Department of Civil and Natural Resources Engineering. He delivers Generative AI training for the 800 students within the department, as well as staff across the wider campus. His research group uses Artificial Intelligence to provide advanced warning of volcanic eruptions, floods and wildfires, while another part of the group works on computational engineering solutions for subsurface energy systems: geothermal, carbon removal, and hydrogen storage. He has previously worked at Los Alamos National Laboratory, Stanford University, and the University of Auckland.
