3 ms·

I think it's because it treats hackers as deficient academics who can't be bothered to cite their sources properly or do any research, while there's an equal an

by strken 9y ago

I think it's because it treats hackers as deficient academics who can't be bothered to cite their sources properly or do any research, while there's an equal and opposite argument that academics are deficient hackers who can't be bothered to put their code on github or write any documentation. In reality, neither is really deficient, they're just aiming at different audiences which need different things.

togelius 9y ago
Oh yes, I'm a terrible hacker, I rarely write code anymore. An unfortunate byproduct of academia is that you get "promoted out of the job", and all the hands-on work is done by your lab members. I didn't mean to imply that tinkering was inferior to research - the whole premise was just to tease out how they're different, with different audiences, as you say. Interestingly, the discussion here has been dominated by people who think that I look down on them. People who've discussed it in other fora have not read the post that way.
- deepnet 9y ago
  Togelius I believe you are in error in Seth Bling's case. The Show More links under his youtube MarI/O video link directly to the Stanley Miikkulainen NEAT paper. and he links the wikipedia page on Neuroevolution https://en.wikipedia.org/wiki/Neuroevolution https://en.wikipedia.org/wiki/Neuroevolution which cites two [1][2] of your papers among others : yet you say : " it is atrocious because of the complete lack of scholarship. The guy didn't even know he was reinventing the wheel, and didn't care to look it up. " a bit harsh perhaps ? At least mentioning the foundational NEAT paper is not a complete lack of scholarship. I feel you have overlooked that Seth does try and provide good citations to his audience and it would kind if you mentioned him by name in your article rather than " some guy " He did inspire other twitch streamers to experiment with neuroevolution and neural nets and benchmark many SNES games.[3] although downstream this follow up does goes uncited by yourself. [1] "Neuroevolution in Games: State of the Art and Open Challenges" https://arxiv.org/pdf/1410.7326v3.pdf https://arxiv.org/pdf/1410.7326v3.pdf [2] "Countering poisonous inputs with memetic neuroevolution" (PDF)https://www.academia.edu/download/30945872/poison.pdf https://www.academia.edu/download/30945872/poison.pdf [3] Mario Bros mari/o https://clips.twitch.tv/CourteousEmpathicTruffleCeilingCat https://clips.twitch.tv/CourteousEmpathicTruffleCeilingCat https://www.youtube.com/watch?v=bRxUQNFxAWc https://www.youtube.com/watch?v=bRxUQNFxAWc Mari/o kart winterbunny https://clips.twitch.tv/HilariousPolishedZucchiniHassanChop https://clips.twitch.tv/HilariousPolishedZucchiniHassanChop - mario kart RNN mariflow

by strken 9y ago

togelius 9y ago

Oh yes, I'm a terrible hacker, I rarely write code anymore. An unfortunate byproduct of academia is that you get "promoted out of the job", and all the hands-on work is done by your lab members. I didn't mean to imply that tinkering was inferior to research - the whole premise was just to tease out how they're different, with different audiences, as you say. Interestingly, the discussion here has been dominated by people who think that I look down on them. People who've discussed it in other fora have not read the post that way.

deepnet 9y ago
Togelius I believe you are in error in Seth Bling's case. The Show More links under his youtube MarI/O video link directly to the Stanley Miikkulainen NEAT paper. and he links the wikipedia page on Neuroevolution https://en.wikipedia.org/wiki/Neuroevolution https://en.wikipedia.org/wiki/Neuroevolution which cites two [1][2] of your papers among others : yet you say : " it is atrocious because of the complete lack of scholarship. The guy didn't even know he was reinventing the wheel, and didn't care to look it up. " a bit harsh perhaps ? At least mentioning the foundational NEAT paper is not a complete lack of scholarship. I feel you have overlooked that Seth does try and provide good citations to his audience and it would kind if you mentioned him by name in your article rather than " some guy " He did inspire other twitch streamers to experiment with neuroevolution and neural nets and benchmark many SNES games.[3] although downstream this follow up does goes uncited by yourself. [1] "Neuroevolution in Games: State of the Art and Open Challenges" https://arxiv.org/pdf/1410.7326v3.pdf https://arxiv.org/pdf/1410.7326v3.pdf [2] "Countering poisonous inputs with memetic neuroevolution" (PDF)https://www.academia.edu/download/30945872/poison.pdf https://www.academia.edu/download/30945872/poison.pdf [3] Mario Bros mari/o https://clips.twitch.tv/CourteousEmpathicTruffleCeilingCat https://clips.twitch.tv/CourteousEmpathicTruffleCeilingCat https://www.youtube.com/watch?v=bRxUQNFxAWc https://www.youtube.com/watch?v=bRxUQNFxAWc Mari/o kart winterbunny https://clips.twitch.tv/HilariousPolishedZucchiniHassanChop https://clips.twitch.tv/HilariousPolishedZucchiniHassanChop - mario kart RNN mariflow