ErgoAI FAQ

[Michael Kifer]

We now have a FAQ for ErgoAI. Please suggest additional topics to include.

Related:

• ErgoAI documentation
• Flora-2 documentation
• XSB documentation

[Paul Payne]

Architectural examples for how one would use ErgoAI in a more comprehensive system. For example, examples of any of these systems would be neat:

- Multi-robot planning.

- Event-reactive systems which react in real-time to event feeds.

- Hybrid AI systems --- suggestions on how to integrate ML.

- SmartContracts -- integrating/replacing Solidity, e.g. or other blockchain-based smart-contract languages.

- A Slack bot.

- A IoT home automation example, perhaps w/ integration w/ Alexa.

There are some low-level examples of, e.g. using SQL. But these higher-level examples would really give lift to projects starting up.

[Michael Kifer]

Yes, Paul,

these are all great examples of applications of reasoning engines.

We are always on a lookout for users who develop interesting applications with Ergo and Flora-2, and we are willing to support such users with advice.

--

       --- michael

--
You received this message because you are subscribed to the Google Groups "ErgoAI, Flora-2, and XSB Users Forum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ErgoAI-Flora2-XSB...@coherentknowledge.com.
To view this discussion on the web visit https://groups.google.com/a/coherentknowledge.com/d/msgid/ErgoAI-Flora2-XSB-forum/13cbd364-641a-41ee-8e8e-17c81ef10d8d%40coherentknowledge.com.

[Robert Goldman]

One thing that confused me a bit was that the aggregators don't work as I expected from Prolog.  I.e., they are invoked as `?Var = aggregator{...}` instead of having the "return value" in the aggregator form.  It's only in the "Here are a few typical uses of aggregation" that this is visible, and there is nothing in the text that draws the reader's attention to this.  A head's up in the topic paragraph might be helpful.

[Andrew Dougherty]

Hi Paul and everyone!

I apologize for the intrusion, however, I have several of the kinds of systems for which you are looking, but currently they are pure (SWI) Prolog. I have worked to integrate my projects with Flora-2 in the past, but now, with Claude's help, I will hopefully integrate with ErgoAI RSN. I have a new mini pc dedicated for these kinds of LLM-driven development efforts arriving today finally.

- Multi-robot planning - Not really, although I have integrated Thielscher's FLUX system, and recently tried with Claude's help to implement a handwavy version of TeamLog recently (since I do not have the skill to fully implement).
- Event-reactive systems which react in real-time to event feeds. - I believe so, I am using a SWI Prolog AgentSpeak(L) Meta-interpreter - https://github.com/aindilis/autonomous-ai-agent/ (subset of Agent-code only - the whole system is fixed-up version of this: https://github.com/aindilis/agentspeak-l-swi-prolog-meta-interpreter ) There is also externally https://github.com/aartikis/RTEC
- Hybrid AI systems --- suggestions on how to integrate ML. - Does LLM integration count?
- SmartContracts -- integrating/replacing Solidity, e.g. or other blockchain-based smart-contract languages. - Not yet, although I've been looking just today at this: https://github.com/triska/bitcoinolog and earlier at https://tau.net/ and autonomic and such.
- A Slack bot. - no (although some Discord integration and tooling).
- A IoT home automation example, perhaps w/ integration w/ Alexa. - Yes, have Alexa and Rhasspy integration working. More information here: https://github.com/aindilis/flp/blob/main/ReferenceManual.md

The catch- the system is currently air gapped behind a hardware-level data-diode, and extremely difficult to release due to being infused without personal (life planning) information throughout. I'm hoping to rapidly reboot these projects using TDD and release early/often with Claude's help and some more manual redaction of existing tools as necessary, as I pursue developing a full-on scrubbing and redaction pipeline to release the whole system (in maybe 2 years, idk - real life).

If you or anyone else would like to collaborate on porting (likely with Claude etc's help) to XSB and integrating some of these tools with ErgoAI that would be amazing. Because I am rebooting the projects, we could think about making the integration more intentional than it was in the last iteration.

Kind Regards,

Andrew

You received this message because you are subscribed to the Google Groups "ErgoAI and XSB Users Forum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ErgoAI-XSB-for...@coherentknowledge.com.
To view this discussion visit https://groups.google.com/a/coherentknowledge.com/d/msgid/ErgoAI-XSB-forum/8e98ab55-68d7-4169-98ed-a1523979d8d9n%40coherentknowledge.com.

--

Andrew Dougherty

Independent AI Researcher

https://frdcsa.org/~andrewdo

(331) 684-7674‬

adou...@gmail.com

[Michael Kifer]

Thanks, Robert. In ergo the aggregates are functions that get evaluated dduring the execution and the resulting values then replace those aggregate expressions.  In other terms, aggregate fincs are, literally,, functions.

The manual does mention this fact, but I added this explanation also in  section 30.1.

 The bottom line is that Ergo's noation is more natural and flexible,