Hacker News Reader

Comment by antfarm | Hacker News Reader

There used to be another open-source agentframework by the same name, but it was for multi-agent simulations. For a moment I thought there was a new wave of interest in a deeper understanding of complex systems by means of modelling.

https://en.wikipedia.org/wiki/Swarm_(simulation)

https://www.santafe.edu/research/results/working-papers/the-...

View on HN

Hey, I wrote that! But it was nearly 30 years ago, it's OK for someone else to use the same name.

Fun fact: Swarm was one of the very few non-NeXT/Apple uses of Objective C. We used the GNU Objective C runtime. Dynamic typing was a huge help for multiagent programming compared to C++'s static typing and lack of runtime introspection. (Again, nearly 30 years ago. Things are different now.)

Hey, thanks for writing the original Swarm! Also thought of that immediately when I saw the headline.

I enjoyed using it around 2002, got introduced via Rick Riolo at the the University of Michigan Center for the Study of Complex Systems. It was a bit of a gateway drug for me from software into modeling, particularly since I was already doing OS X/Cocoa stuff in Objective-C.

A lot of scientific modelers start with differential equations, but coming from object-oriented software ABMs made a lot more sense to me, and learning both approaches in parallel was really helpful in thinking about scale, dimensionality, representation, etc. in the modeling process, as ODEs and complex ABMs—often pathologically complex—represent end points of a continuum.

Tangentially, in one of Rick's classes we read about perceptrons, and at one point the conversation turned to, hey, would it be possible to just dump all the text of the Internet into a neural net? And here we are.

I took a graduate level class in the 1990s from some SFI luminaries. It was a great class but the dismal conclusion was "this stuff is kind of neat but not very practical, traditional optimization techniques usually work better". None of us guessed if you could scale the networks up 1 million X or more they'd become magic.

> compared to C++'s static typing and lack of runtime introspection. (Again, nearly 30 years ago. Things are different now.)

C++ has added a ton of great features since (especially C++11 onward) but run-time reflection is still sorely missed.

Hey thanks for writing the original swarm. I found your framework very inspiring when I was conducting my own personal (pretty much universally failed) experiments into making this kind of multi-agent simulation.

I believe there is a new wave of interest in deeper understanding of complex systems through modelling and connecting with machine learning. I organized this conference on exploring system dynamics with AI which you can see most of the lectures here:

https://youtube.com/playlist?list=PL6zSfYNSRHalAsgIjHHsttpYf...

The idea was to think about it from different directions including academia, industry, and education.

Nobody presented multi agent simulations but I agree with you that is a very interesting way of thinking about things. There was a talk on high dimensional systems modelled with networks but the speaker didn't want their talk published online.

Anyways I'm happy to chat more about these topics. I'm obsessed with understanding complexity using ai, modelling, and other methods.

An AI conference that isn't bullshit hype? Will wonders never cease?

> Nobody presented multi agent simulations but I agree with you that is a very interesting way of thinking about things.

To answer your question I did build a simulation of how a multi model agent swarm - agents have different capabilities and run times - would impact the end user wait time based on arbitrary message parsing graphs.

After playing with it for an afternoon I realized I was basically doing a very wasteful Markov chain enumeration algorithm and wrote one up accordingly.

Yeah I already have loads of people asking when the next one is for this exact reason. Haha. Well, I would love to have people help organise the next one. But I don't know yet

This looks rad! But you should title the videos with the topic and the speakers name, and if you must include the conference name, put it at the end :)

As-is, it's hard to skim the playlist, and likely terrible for organic search on Google or YouTube <3

I agree with you. Unfortunately I'm not in charge of the videos so even though I asked them to do this they didn't. Haha.

Reply on news.ycombinator.com