Ken Stanley: Discovery Without Objectives
Channel:fzambetta75 · Length:1:32:20 · Published:2012-09-04 · Views:5662
Summary
A 92-minute RMIT University seminar in which Ken Stanley argues that setting an objective can block achieving it. He presents objective-less, novelty-driven search as an alternative approach to innovation and discovery, drawing on evolutionary computation.
Why HN recommends it
Cited as a source for the idea that open-ended search can beat objective-driven optimization, though one commenter found parts of it fluffy.
I've seen a couple of really interesting talks recently about the surprising successfulness of objective-less search by Ken Stanley — HN, 2018
Machine learning & deep learning · Evolution & genetics · Scientific method & science communication
Subtitles (excerpt)
Good evening ladies and gentlemen. Welcome to the big picture seminar series. My name is Lana. I'm the external relations manager Victoria Research Lab. So it's my great pleasure to welcome you here this evening. The purpose of the big future seminar is really to provide the public with an opportunity to hear from people who are shaping our technology future. And this evening we we're very pleased to hold this event here at RMIT University. RMIT is a part of the University of NICTA here in in Melbourne. So we're very pleased to have the opportunity to uh uh provide this uh seminar here at the university. Uh it's my uh pleasure to invite Dr. Fabio Zeta from the school of computer science and IT at RIT University to introduce our speaker. Please welcome Fabio. Good evening everybody. um is my great pleasure. I'm here can sting uh associate professor um part of computer science UCF universal university university of central Florida um where he heads the uh elex evolutionary complexity research group. Um he's going to talk about discovery without objectives tonight. Uh we're very pleased to have him here. Um I'd like to acknowledge support from our great sponsors um of course and uh as well as ACM through the distinguished lecture uh program.
On Hacker News
I don't think this algorithm is very common. In particular, the colour palette can usually be chosen last , and a lot of generative art (at least all that shown in the article) will not need multiple resolution passes. I think a more common discovery/refinement process is to go from more random to less random, or random in particular kinds of ways. For example, you can specify the values or distr… — mkl, 2018-09
Fascinating. Instead of banning money or fixing prices (did they try that in Russia?) we retain free markets but now with a new layer of government intervention. Not intervention by social spending or regulation but by buying and selling according to the dictats of an AI. And the utility function of this AI is set according to what exactly? My guess is it would have to be according to our values,… — truculation, 2018-03
Interesting talk. I really couldn't tell how much of it really meant something and how much of it was fluff like "power poses". In any case he's been giving similar talks for a while. [VIDEO] — greggman, 2016-02
Related videos
- Jackie Chan - How to Do Action Comedy · Every Frame a Painting · 9:04
- The spelled-out intro to neural networks and backpropagation: building micrograd · Andrej Karpathy · 2:25:52
- The Quantum Conspiracy: What Popularizers of QM Don't Want You to Know · Google TechTalks · 1:03:43
- But what is a neural network? | Deep learning chapter 1 · 3Blue1Brown · 18:40
- James Burke Connections, Ep. 1 "The Trigger Effect" · DRS_Education · 49:22
- AlphaGo - The Movie | Full award-winning documentary · Google DeepMind · 1:30:28
- Transformers, the tech behind LLMs | Deep Learning Chapter 5 · 3Blue1Brown · 27:14
- USENIX Security '18-Q: Why Do Keynote Speakers Keep Suggesting That Improving Security Is Possible? · USENIX · 51:22
- Lecture 1 | Machine Learning (Stanford) · Stanford · 1:08:40
- Hamming, "You and Your Research" (June 6, 1995) · securitylectures · 44:02
- MarI/O - Machine Learning for Video Games · SethBling · 5:57
- Deep Dive into LLMs like ChatGPT · Andrej Karpathy · 3:31:23
- Computer program that learns to play classic NES games · suckerpinch · 16:17
- Why Being Delusional is a Superpower · Veritasium · 12:04
- RL Course by David Silver - Lecture 1: Introduction to Reinforcement Learning · Google DeepMind · 1:28:12
- Is Most Published Research Wrong? · Veritasium · 12:22