Archive - The Agentic Edge 🔀

October 2024

Agents and constraints

Don't restrict your agents unless you absolutely have to.

Oct 24 •

Simpson's Paradox

When the statistics of the parts and the whole don't tell the same story

Oct 24

Your CMS is a gold mine: time to dig

The future of content management and personalization powered by AI

Oct 23 •

and

Arpit Choudhury

42:40

Agents, impact confidence, and prior beliefs

Agents might use data, but they need prior beliefs too, just like humans do

Oct 17 •

6:27

What is an AI Agent? Definitions, Inputs, and Outputs

AI Agents don't *have* to be LLM-based

Oct 9 •

Arpit Choudhury

and

Embedding space let agents communicate with each other

Team members need to be able to coordinate and swap notes. Even if the team is made up entirely of virtual agents.

Oct 8 •

5:22

Get comfortable with Decision Automation

Explore the complexities of automating business decisions for improved decision-making

Oct 7 •

Arpit Choudhury

,

Paul Meinshausen

, and

37:19

Transforming distributions to encode business priorities

Agents need to navigate both the explore/exploit and the user/business trade-offs.

Oct 4 •

5:04

Beta distribution draws and signal strength

A probability is not a point estimate

Oct 4 •

5:37

How agents view behavioral signals

Use an adaptation of Interrupted Time Series analysis

Oct 4 •

4:24

#nojs-banner { position: fixed; bottom: 0; left: 0; padding: 16px 16px 16px 32px; width: 100%; box-sizing: border-box; background: red; color: white; font-family: -apple-system, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol"; font-size: 13px; line-height: 13px; } #nojs-banner a { color: inherit; text-decoration: underline; } This site requires JavaScript to run correctly. Please turn on JavaScript or unblock scripts