The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities · HackerPapers discussions

The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities — mentioned by sbierwagen

Thu, 01 Jun 2023 23:51:01 GMT

Spiritually related: examples of RL algorithms exploiting glitches in the physics engine (clipping into the ground to shoot forward at incredible speed) or gaming the goal state: (Agent that's supposed to learn how to walk instead evolving a body with a single long leg that just falls forward-- getting a high speed score with low complexity penalty) https://arxiv.org/abs/1803.03453 https://www.deepmind.com/blog/specification-gaming-the-flip-...

Paper and discussion history. A mention is not an endorsement.

The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities — mentioned by YeGoblynQueenne

Wed, 05 Apr 2023 10:24:05 GMT

Again, you're talking about unpredictability and the inability to precisely trace the execution of a complex procedure, not about inability to understand how something works, or its general capabilities. Like I say in my other comments above, that a system is unpredictable isn't reason to assume that it's doing something it wasn't designed to do. Not when the system in question is a computer system, or a program, and we know how it was made and why (and given that the unpredictable behaviour is not the result of bugs; I don't think anyone is saying that the "sparks of AGI" in LLMs are the…

Paper and discussion history. A mention is not an endorsement.

The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities — mentioned by cmorez

Sat, 20 Aug 2022 12:52:44 GMT

> I don't understand how you can look at the current or even predicted state of the technology that we have and say "we are nowhere near the point where controlling an AGI is possible". Like....just pull the plug. https://www.deepmind.com/blog/specification-gaming-the-flip-... https://vkrakovna.wordpress.com/2018/04/02/specification-gam... https://arxiv.org/abs/1803.03453 (The Surprising Creativity of Digital Evolution) We literally don't know how to stop effective optimizing processes, deployed in non-handcrafted environments, from discovering solutions and workarounds that satisfy the…

Paper and discussion history. A mention is not an endorsement.

The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities — mentioned by sbierwagen

Tue, 28 Sep 2021 21:24:29 GMT

Reinforcement learning bots exploiting physics engine bugs has a long and illustrious history: https://arxiv.org/pdf/1803.03453.pdf https://news.ycombinator.com/item?id=16600701

Paper and discussion history. A mention is not an endorsement.

The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities — HN submission

Tue, 06 Jul 2021 16:24:50 GMT

A collection of anecdotes on AI oddities

1 point; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities — mentioned by AceJohnny2

Tue, 06 Jul 2021 16:24:20 GMT

There are many such stories of AI "optimizations" gone wrong, because of loopholes the program found that humans didn't consider. Here's a collection of such stories: https://arxiv.org/pdf/1803.03453.pdf

Paper and discussion history. A mention is not an endorsement.

The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities — mentioned by versteegen

Sat, 26 Jun 2021 08:51:57 GMT

This gave me a realisation: manipulative (deceptive) artificial life and artificial intelligence are forms of parasites. But instead of manipulating/deceiving their host organisms, they deceive the humans who control the life simulation/the environment the AI exists within. For example, if a human controller is trying to study the evolution of certain artificial lifes by filtering others out of the simulated world based on their behaviour, what happens? Life evolves to detect, if possible, when they are being observed and act differently in that situation to deceive the human observer. See…

Paper and discussion history. A mention is not an endorsement.

The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities — mentioned by Matumio

Fri, 11 Jun 2021 20:20:51 GMT

There is a 2018 paper called "The Surprising Creativity of Digital Evolution" which is a collection of similar anecdotes: https://arxiv.org/abs/1803.03453

Paper and discussion history. A mention is not an endorsement.

The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities — mentioned by gwern

Sun, 07 Mar 2021 00:42:21 GMT

That might be the lexicographic Mario agent: http://tom7.org/mario/ (Checking my list & https://arxiv.org/abs/1803.03453 I don't see any others about pausing or stopping the game.)

Paper and discussion history. A mention is not an endorsement.

The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities — mentioned by fsiefken

Tue, 09 Jun 2020 09:21:18 GMT

I was wondering if there was any work in the field of evolving 'creatures' moving in a virtual space, beyond what Karl Sims did 23 years ago. https://www.karlsims.com/evolved-virtual-creatures.html https://www.karlsims.com/galapagos/index.html It would be nice if Karl Sims could open source it as it is really inspiring visual example in the field of Artificial Life, next to seeing the generations of metuselahs unfolding in Conway's game of life Another interesting article in the field of Artificial Life: https://arxiv.org/pdf/1803.03453.pdf

Paper and discussion history. A mention is not an endorsement.

The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities — HN submission

Tue, 26 May 2020 09:01:17 GMT

The Surprising Creativity of Digital Evolution

2 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities — HN submission

Wed, 21 Aug 2019 04:42:09 GMT

The Surprising Creativity of Digital Evolution

2 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities — mentioned by Bartweiss

Thu, 25 Apr 2019 21:11:41 GMT

Oh man, good question. I'm always up for swapping these stories. A lot of these came from a paper on weird AI tricks, and resulting best-of list on a blog collecting these stories.[1][2] Suffice to say, the people who think the orthogonality thesis is a weird hypothetical aren't keeping up with the state of things. - The aforementioned Tetris story: an undirected learner was set to maximize score at Tetris learned normal play techniques, but also learned to pause the game immediately before losing so that the score wouldn't "decline" at game over. - In the same vein as interns quitting, proxy…

Paper and discussion history. A mention is not an endorsement.

The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities — mentioned by YeGoblynQueenne

Tue, 22 Jan 2019 19:04:04 GMT

I see what you mean- thank you for the clarification. Well, I'm not really the right person to discuss this issue since I can't claim to understand how the brain works. However, I do understand a few things about connectionist methods and it's my understanding that they are not very good models of the way the brain works at all (am I misrepresenting your turn of phrase, of "biologically plausible"?). In that sense, I doubt it's possible that a neural net would develop symbolic reasoning by dint of being in some way similar to a biological brain. In general, my experience with neural nets and…

Paper and discussion history. A mention is not an endorsement.

The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities — mentioned by carapace

Thu, 15 Nov 2018 19:47:57 GMT

Cf. "The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities" https://arxiv.org/abs/1803.03453v1 > Biological evolution provides a creative fount of complex and subtle adaptations, often surprising the scientists who discover them. However, because evolution is an algorithmic process that transcends the substrate in which it occurs, evolution's creativity is not limited to nature. Indeed, many researchers in the field of digital evolution have observed their evolving algorithms and organisms…

Paper and discussion history. A mention is not an endorsement.

The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities — mentioned by bediger4000

Wed, 14 Nov 2018 15:51:56 GMT

Sure. Why is the title of the article so disingenuous? What design do people think they see in nature? By what method can I, without referring to someone else's authority, decide that a particular part of nature is "designed"? Without an explicit criteria for "design", the whole thing is just arguing over opinions about what is designed and what isn't. That is, it would become just like Behe's "irreducible complexity". No definite criteria for "irreducible" and no measurement for "complexity". While you're at it, point me to some PDFs of refereed papers written by Dembski. I'm a smart guy, I…

Paper and discussion history. A mention is not an endorsement.

The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities — mentioned by arctop_dan

Mon, 08 Oct 2018 01:53:20 GMT

I was forwarded this discussion by a friend who’s familiar with my work and I’ve really been enjoying the posts. I think about ethics and AI a lot, and couldn’t help but want to contribute a few thoughts here. So here it goes, my first post on HN.. My main advice is beware of AI’s surprising creativity and proactively work to insure it stays aligned with human interest. There was a fascinating crowd-sourced paper published this year that shares anecdotes about unexpected adaptations encountered by researchers working in artificial life and evolutionary computation[1]. These are the sort of…

Paper and discussion history. A mention is not an endorsement.

The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities — mentioned by rhaps0dy

Fri, 01 Jun 2018 20:14:07 GMT

I'm not so sure. Even prosaic, current AI, can be very unpredictable. See this blog post http://aiweirdness.com/post/172894792687/when-algorithms-sur... , which is a somewhat more causal summary of this paper "The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities" https://arxiv.org/abs/1803.03453v2 The authors of the paper explain a bunch of anecdotes when they wanted the AI to do one thing, but the measure they told it to optimise for wasn't actually what they intended, and so unexpected things…

Paper and discussion history. A mention is not an endorsement.

The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities — mentioned by matthberg

Fri, 25 May 2018 10:23:10 GMT

This a pretty much direct summary of a paper [0] previously discussed(twice) on hacker news [1][2] [0]: https://arxiv.org/pdf/1803.03453.pdf [1]: https://news.ycombinator.com/item?id=16837030 [2]: https://news.ycombinator.com/item?id=16600701

Paper and discussion history. A mention is not an endorsement.

The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities — mentioned by pietroglyph

Sat, 14 Apr 2018 22:31:12 GMT

This article is actually providing select anecdotes from a more exhaustive paper: https://arxiv.org/pdf/1803.03453.pdf

Paper and discussion history. A mention is not an endorsement.

The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities — HN submission

Sat, 14 Apr 2018 12:31:23 GMT

When algorithms surprise us

112 points; 18 discussion replies on HN. Paper and source history. A submission is not an endorsement.

The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities — mentioned by ericjang

Thu, 22 Mar 2018 16:34:01 GMT

Here's a recent survey / observational science paper by some prominent "neuroevolution" / A-Life researchers. https://arxiv.org/abs/1803.03453. I found this refreshing because it's rare that science papers talk about the debugging and experimental process and debugging journeys underlying this research.

Paper and discussion history. A mention is not an endorsement.

The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities — HN submission

Fri, 16 Mar 2018 14:05:27 GMT

Surprising Creativity: Anecdotes from Evolutionary Computation

86 points; 19 discussion replies on HN. Paper and source history. A submission is not an endorsement.

The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities — HN submission

Thu, 15 Mar 2018 13:54:34 GMT

The Surprising Creativity of Digital Evolution: Anecdotes from AI Researchers

1 point; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities — HN submission

Tue, 13 Mar 2018 16:26:14 GMT

Evolutionary AI routinely outsmarts the scientists that use it

1 point; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities — HN submission

Tue, 13 Mar 2018 07:20:53 GMT

The Surprising Creativity of Digital Evolution

2 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities — HN submission

Mon, 12 Mar 2018 19:46:54 GMT

The Surprising Creativity of Digital Evolution: Artifical Life and EC Anecdotes

2 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.