Mastodon

Teixi<a href="https://mastodon.social/tags/ACMPrize" class="mention hashtag" rel="nofollow noopener" target="_blank">#ACMPrize</a> <a href="https://mastodon.social/tags/2024ACMPrize" class="mention hashtag" rel="nofollow noopener" target="_blank">#2024ACMPrize</a> <a href="https://mastodon.social/tags/ACMTuringAward" class="mention hashtag" rel="nofollow noopener" target="_blank">#ACMTuringAward</a><a href="https://mastodon.social/tags/AndrewBarto" class="mention hashtag" rel="nofollow noopener" target="_blank">#AndrewBarto</a> <a href="https://mastodon.social/tags/RichardSutton" class="mention hashtag" rel="nofollow noopener" target="_blank">#RichardSutton</a> » <a href="https://mastodon.social/tags/ReinforcementLearning" class="mention hashtag" rel="nofollow noopener" target="_blank">#ReinforcementLearning</a> An Introduction 1998 standard reference...cited over 75,000 ... prominent example of <a href="https://mastodon.social/tags/RL" class="mention hashtag" rel="nofollow noopener" target="_blank">#RL</a> <a href="https://mastodon.social/tags/AlphaGo" class="mention hashtag" rel="nofollow noopener" target="_blank">#AlphaGo</a> victory over best human <a href="https://mastodon.social/tags/Go" class="mention hashtag" rel="nofollow noopener" target="_blank">#Go</a> players 2016 2017 .... recently has been the development of the chatbot <a href="https://mastodon.social/tags/ChatGPT" class="mention hashtag" rel="nofollow noopener" target="_blank">#ChatGPT</a> ... large language model <a href="https://mastodon.social/tags/LLM" class="mention hashtag" rel="nofollow noopener" target="_blank">#LLM</a> trained in two phases ...employs a technique called reinforcement learning from human feedback <a href="https://mastodon.social/tags/RLHF" class="mention hashtag" rel="nofollow noopener" target="_blank">#RLHF</a> «aka cheap labor unnamed in papers<a href="https://awards.acm.org/about/2024-turing" rel="nofollow noopener" translate="no" target="_blank">https://awards.acm.org/about/2024-turing</a>2/2

Hacker NewsReflection – AlphaGo / Gemini team building superintelligent coding agents — <a href="https://www.reflection.ai/superintelligence/" rel="nofollow noopener" translate="no" target="_blank">https://www.reflection.ai/superintelligence/</a> <a href="https://mastodon.social/tags/HackerNews" class="mention hashtag" rel="nofollow noopener" target="_blank">#HackerNews</a> <a href="https://mastodon.social/tags/Reflection" class="mention hashtag" rel="nofollow noopener" target="_blank">#Reflection</a> <a href="https://mastodon.social/tags/AlphaGo" class="mention hashtag" rel="nofollow noopener" target="_blank">#AlphaGo</a> <a href="https://mastodon.social/tags/Gemini" class="mention hashtag" rel="nofollow noopener" target="_blank">#Gemini</a> <a href="https://mastodon.social/tags/Superintelligence" class="mention hashtag" rel="nofollow noopener" target="_blank">#Superintelligence</a> <a href="https://mastodon.social/tags/CodingAgents" class="mention hashtag" rel="nofollow noopener" target="_blank">#CodingAgents</a> <a href="https://mastodon.social/tags/AIResearch" class="mention hashtag" rel="nofollow noopener" target="_blank">#AIResearch</a>

LavX NewsReinforcement Learning Revolution: Turing Prize Winners Shape AI's FutureThe recent Turing Award honors Andrew G. Barto and Richard S. Sutton for their groundbreaking work in reinforcement learning, a technique that has become essential in modern AI applications. Their con...<a href="https://news.lavx.hu/article/reinforcement-learning-revolution-turing-prize-winners-shape-ai-s-future" rel="nofollow noopener" target="_blank">https://news.lavx.hu/article/reinforcement-learning-revolution-turing-prize-winners-shape-ai-s-future</a><a href="https://mastodon.cloud/tags/news" class="mention hashtag" rel="nofollow noopener" target="_blank">#news</a> <a href="https://mastodon.cloud/tags/tech" class="mention hashtag" rel="nofollow noopener" target="_blank">#tech</a> <a href="https://mastodon.cloud/tags/ReinforcementLearning" class="mention hashtag" rel="nofollow noopener" target="_blank">#ReinforcementLearning</a> <a href="https://mastodon.cloud/tags/TuringAward" class="mention hashtag" rel="nofollow noopener" target="_blank">#TuringAward</a> <a href="https://mastodon.cloud/tags/AlphaGo" class="mention hashtag" rel="nofollow noopener" target="_blank">#AlphaGo</a>

Vaughn<a href="https://twit.social/@leo" class="u-url mention" rel="nofollow noopener" target="_blank">@leo</a> I love that you keep mentioning move 37 that AlphaGo made against Lee Sedol. The most poetic thing is that Lee made a similar move in the very next game (the only one he won) that the pros thought was a mistake and that seemed to confuse AlphaGo. <a href="https://m.socialyeti.club/tags/go" class="mention hashtag" rel="nofollow noopener" target="_blank">#go</a> <a href="https://m.socialyeti.club/tags/alphago" class="mention hashtag" rel="nofollow noopener" target="_blank">#alphago</a> <a href="https://m.socialyeti.club/tags/baduk" class="mention hashtag" rel="nofollow noopener" target="_blank">#baduk</a> <a href="https://m.socialyeti.club/tags/weiqi" class="mention hashtag" rel="nofollow noopener" target="_blank">#weiqi</a>

Socied@d ReticularPequeños y grandes pasos hacia el imperio de la inteligencia artificial<a href="https://mas.to/@echo_xc@mastodon.social/113971286060501063" rel="nofollow noopener" target="_blank"></a>Fuente: Open TechTraducción de la infografía:<ul><li>1943 – McCullock y Pitts publican un artículo titulado Un cálculo lógico de ideas inmanentes en la actividad nerviosa, en el que proponen las bases para las redes neuronales.</li></ul><ul><li>1950 – Turing publica Computing Machinery and Intelligence, proponiendo el Test de Turing como forma de medir la capacidad de una máquina.</li></ul><ul><li>1951 – Marvin Minsky y Dean Edmonds construyen SNAR, la primera computadora de red neuronal.</li></ul><ul><li>1956 – Se celebra la Conferencia de Dartmouth (organizada por McCarthy, Minsky, Rochester y Shannon), que marca el nacimiento de la IA como campo de estudio.</li></ul><ul><li>1957 – Rosenblatt desarrolla el Perceptrón: la primera red neuronal artificial capaz de aprender.</li></ul>(!!) Test de Turing: donde un evaluador humano entabla una conversación en lenguaje natural con una máquina y un humano.<ul><li>1965 – Weizenbaum desarrolla ELIZA: un programa de procesamiento del lenguaje natural que simula una conversación.</li></ul><ul><li>1967 – Newell y Simon desarrollan el Solucionador General de Problemas (GPS), uno de los primeros programas de IA que demuestra una capacidad de resolución de problemas similar a la humana.</li></ul><ul><li>1974 – Comienza el primer invierno de la IA, marcado por una disminución de la financiación y del interés en la investigación en IA debido a expectativas poco realistas y a un progreso limitado.</li></ul><ul><li>1980 – Los sistemas expertos ganan popularidad y las empresas los utilizan para realizar previsiones financieras y diagnósticos médicos.</li></ul><ul><li>1986 – Hinton, Rumelhart y Williams publican Aprendizaje de representaciones mediante retropropagación de errores, que permite entrenar redes neuronales mucho más profundas.</li></ul>(!!) Redes neuronales: modelos de aprendizaje automático que imitan el cerebro y aprenden a reconocer patrones y hacer predicciones a través de conexiones neuronales artificiales.<ul><li>1997 – Deep Blue de IBM derrota al campeón mundial de ajedrez Kasparov, siendo la primera vez que una computadora vence a un campeón mundial en un juego complejo.</li></ul><ul><li>2002 – iRobot presenta Roomba, el primer robot aspirador doméstico producido en serie con un sistema de navegación impulsado por IA.</li></ul><ul><li>2011 – Watson de IBM derrota a dos ex campeones de Jeopardy!.</li></ul><ul><li>2012 – La startup de inteligencia artificial DeepMind desarrolla una red neuronal profunda que puede reconocer gatos en vídeos de YouTube.</li></ul><ul><li>2014 – Facebook crea DeepFace, un sistema de reconocimiento facial que puede reconocer rostros con una precisión casi humana.</li></ul>(!!) DeepMind fue adquirida por Google en 2014 por 500 millones de dólares.<ul><li>2015 – AlphaGo, desarrollado por DeepMind, derrota al campeón mundial Lee Sedol en el juego de Go.</li></ul><ul><li>2017 – AlphaZero de Google derrota a los mejores motores de ajedrez y shogi del mundo en una serie de partidas.</li></ul><ul><li>2020 – OpenAI lanza GPT-3, lo que marca un avance significativo en el procesamiento del lenguaje natural.</li></ul>(!!) Procesamiento del lenguaje natural: enseña a las computadoras a comprender y utilizar el lenguaje humano mediante técnicas como el aprendizaje automático.<ul><li>2021 – AlphaFold2 de DeepMind resuelve el problema del plegamiento de proteínas, allanando el camino para nuevos descubrimientos de fármacos y avances médicos.</li></ul><ul><li>2022 – Google despide al ingeniero Blake Lemoine por sus afirmaciones de que el modelo de lenguaje para aplicaciones de diálogo (LaMDA) de Google era sensible.</li></ul><ul><li>2023 – Artistas presentaron una demanda colectiva contra Stability AI, DeviantArt y Mid-journey por usar Stable Diffusion para remezclar las obras protegidas por derechos de autor de millones de artistas.</li></ul>Gráfico: <a href="https://mas.to/@echo_xc@mastodon.social/113971286060501063" rel="nofollow noopener" target="_blank">Open Tech</a> / <a href="https://www.genuineimpact.io/" rel="nofollow noopener" target="_blank">Genuine Impact</a>Entradas relacionadas<ul><li><a href="https://anselmolucio.wordpress.com/2025/01/31/como-definir-la-credibilidad-algoritmica-deepseek-da-en-el-clavo/" rel="nofollow noopener" target="_blank">¿Cómo definir la «credibilidad algorítmica»? DeepSeek da en el clavo</a></li><li><a href="https://anselmolucio.wordpress.com/2024/12/16/los-precios-dinamicos-exacerban-la-desigualdad-entre-los-consumidores-hay-que-regularlos-ya/" rel="nofollow noopener" target="_blank">Los precios dinámicos exacerban la desigualdad entre los consumidores, hay que regularlos ya</a></li><li><a href="https://anselmolucio.wordpress.com/2021/02/22/firma-contra-la-vigilancia-biometrica-masiva/" rel="nofollow noopener" target="_blank">Firma contra la vigilancia biométrica masiva</a></li></ul><a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/ajedrez/" target="_blank">#ajedrez</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/alphafold2/" target="_blank">#AlphaFold2</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/alphago/" target="_blank">#AlphaGo</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/alphazero/" target="_blank">#AlphaZero</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/aprendizaje-automatico/" target="_blank">#aprendizajeAutomático</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/articulo/" target="_blank">#artículo</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/artistas/" target="_blank">#artistas</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/aspirador/" target="_blank">#aspirador</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/blake-lemoine/" target="_blank">#BlakeLemoine</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/conferencia-de-dartmouth/" target="_blank">#ConferenciaDeDartmouth</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/copyright/" target="_blank">#copyright</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/dean-edmonds/" target="_blank">#DeanEdmonds</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/deep-blue/" target="_blank">#DeepBlue</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/deepface/" target="_blank">#DeepFace</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/deepmind/" target="_blank">#DeepMind</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/deviantart/" target="_blank">#DeviantArt</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/eliza/" target="_blank">#ELIZA</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/facebook/" target="_blank">#Facebook</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/gatos/" target="_blank">#gatos</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/genuine-impact/" target="_blank">#GenuineImpact</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/go/" target="_blank">#Go</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/google/" target="_blank">#Google</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/gps/" target="_blank">#GPS</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/gpt-3/" target="_blank">#GPT3</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/grafico/" target="_blank">#gráfico</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/hinton/" target="_blank">#Hinton</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/ia/" target="_blank">#IA</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/ibm/" target="_blank">#IBM</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/infografia/" target="_blank">#infografía</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/inteligencia-artificial/" target="_blank">#inteligenciaArtificial</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/irobot/" target="_blank">#iRobot</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/jeopardy/" target="_blank">#Jeopardy_</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/kasparov/" target="_blank">#Kasparov</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/lamda/" target="_blank">#LaMDA</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/lee-sedol/" target="_blank">#LeeSedol</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/marvin-minsky/" target="_blank">#MarvinMinsky</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/mccarthy/" target="_blank">#McCarthy</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/mccullock/" target="_blank">#McCullock</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/mid-journey/" target="_blank">#MidJourney</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/modelos/" target="_blank">#modelos</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/newell/" target="_blank">#Newell</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/open-tech/" target="_blank">#OpenTech</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/openai/" target="_blank">#OpenAI</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/patrones/" target="_blank">#patrones</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/perceptron/" target="_blank">#Perceptron</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/pitts/" target="_blank">#Pitts</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/plegamiento-de-proteinas/" target="_blank">#plegamientoDeProteínas</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/predicciones/" target="_blank">#predicciones</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/procesamiento-del-lenguaje-natural/" target="_blank">#procesamientoDelLenguajeNatural</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/reconocimiento-facial/" target="_blank">#reconocimientoFacial</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/redes-neuronales/" target="_blank">#redesNeuronales</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/remezclar/" target="_blank">#remezclar</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/robot/" target="_blank">#robot</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/rochester/" target="_blank">#Rochester</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/roomba/" target="_blank">#Roomba</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/rosenblatt/" target="_blank">#Rosenblatt</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/rumelhart/" target="_blank">#Rumelhart</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/shannon/" target="_blank">#Shannon</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/shogi/" target="_blank">#shogi</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/simon/" target="_blank">#Simon</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/sistema-de-navegacion/" target="_blank">#sistemaDeNavegación</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/snar/" target="_blank">#SNAR</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/stability-ai/" target="_blank">#StabilityAI</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/stable-diffusion/" target="_blank">#StableDiffusion</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/test-de-turing/" target="_blank">#testDeTuring</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/turing/" target="_blank">#Turing</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/videos/" target="_blank">#vídeos</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/watson/" target="_blank">#Watson</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/weizenbaum/" target="_blank">#Weizenbaum</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/williams/" target="_blank">#Williams</a> <a rel="nofollow noopener" class="hashtag u-tag u-category" href="https://anselmolucio.wordpress.com/tag/youtube/" target="_blank">#YouTube</a>

martin voggenbergeri doubt that most of humanity has any 'idea' what it would be like to be 'truly' intelligent. and i don’t mean ‘intelligence’ in the context of being clever or doing complex calculations. it takes no true intelligence for that — it’s computational. in the near future, we won’t even need all these clever ones; everything they do can be a automated. the alphago event was a great insight into what's to come.<a href="https://youtu.be/WXuK6gekU1Y?list=LL" rel="nofollow noopener" translate="no" target="_blank">https://youtu.be/WXuK6gekU1Y?list=LL</a><a href="https://mastodon.social/tags/intelligence" class="mention hashtag" rel="nofollow noopener" target="_blank">#intelligence</a> <a href="https://mastodon.social/tags/alphago" class="mention hashtag" rel="nofollow noopener" target="_blank">#alphago</a> <a href="https://mastodon.social/tags/upsidedown" class="mention hashtag" rel="nofollow noopener" target="_blank">#upsidedown</a>

Rod2ik 🇪🇺 🇨🇵 🇪🇸 🇺🇦 🇨🇦 🇩🇰 🇬🇱Le moment <a href="https://mastodon.social/tags/DeepSeek" class="mention hashtag" rel="nofollow noopener" target="_blank">#DeepSeek</a> (2025) est la conséquence du moment <a href="https://mastodon.social/tags/AlphaGo" class="mention hashtag" rel="nofollow noopener" target="_blank">#AlphaGo</a> (2010) de <a href="https://mastodon.social/tags/Google" class="mention hashtag" rel="nofollow noopener" target="_blank">#Google</a> <a href="https://mastodon.social/tags/Deepmind" class="mention hashtag" rel="nofollow noopener" target="_blank">#Deepmind</a> : il a été vécu comme le moment <a href="https://mastodon.social/tags/Spoutnik" class="mention hashtag" rel="nofollow noopener" target="_blank">#Spoutnik</a> (1957) de la <a href="https://mastodon.social/tags/Chine" class="mention hashtag" rel="nofollow noopener" target="_blank">#Chine</a> pour l' <a href="https://mastodon.social/tags/IA" class="mention hashtag" rel="nofollow noopener" target="_blank">#IA</a> <a href="https://mastodon.social/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#AI</a><a href="https://www.numerama.com/tech/1894778-alphago-comment-la-raclee-subie-par-la-chine-au-go-explique-deepseek-aujourdhui.html" rel="nofollow noopener" translate="no" target="_blank">https://www.numerama.com/tech/1894778-alphago-comment-la-raclee-subie-par-la-chine-au-go-explique-deepseek-aujourdhui.html</a>

Rod2ik 🇪🇺 🇨🇵 🇪🇸 🇺🇦 🇨🇦 🇩🇰 🇬🇱Le moment <a href="https://bsky.app/search?q=%23DeepSeek" rel="nofollow noopener" target="_blank">#DeepSeek</a> (2025) est la conséquence du moment <a href="https://bsky.app/search?q=%23AlphaGo" rel="nofollow noopener" target="_blank">#AlphaGo</a> (2010) de <a href="https://bsky.app/search?q=%23Google" rel="nofollow noopener" target="_blank">#Google</a> <a href="https://bsky.app/search?q=%23Deepmind" rel="nofollow noopener" target="_blank">#Deepmind</a> : il a été vécu comme le moment <a href="https://bsky.app/search?q=%23Spoutnik" rel="nofollow noopener" target="_blank">#Spoutnik</a> (1957) de la <a href="https://bsky.app/search?q=%23Chine" rel="nofollow noopener" target="_blank">#Chine</a> pour l' <a href="https://bsky.app/search?q=%23IA" rel="nofollow noopener" target="_blank">#IA</a> <a href="https://bsky.app/search?q=%23AI" rel="nofollow noopener" target="_blank">#AI</a> <a href="https://www.numerama.com/tech/1894778-alphago-comment-la-raclee-subie-par-la-chine-au-go-explique-deepseek-aujourdhui.html" rel="nofollow noopener" target="_blank">www.numerama.com/tech/1894778...</a> <a href="https://www.numerama.com/tech/1894778-alphago-comment-la-raclee-subie-par-la-chine-au-go-explique-deepseek-aujourdhui.html" rel="nofollow noopener" target="_blank">Comment AlphaGo a joué un rôle...</a>

rexi<a href="https://techxplore.com/news/2024-12-ai-human-general-intelligence.html" rel="nofollow noopener" translate="no" target="_blank">https://techxplore.com/news/2024-12-ai-human-general-intelligence.html</a><a href="https://mastodon.social/tags/OpenAI" class="mention hashtag" rel="nofollow noopener" target="_blank">#OpenAI</a> started with a general-purpose version of the <a href="https://mastodon.social/tags/o3system" class="mention hashtag" rel="nofollow noopener" target="_blank">#o3system</a> (which…can spend more time "thinking" about difficult questions) and then trained it specifically for the ARC-AGI test.French <a href="https://mastodon.social/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#AI</a> researcher Francois Chollet…believes o3 searches through different "chains of thought" describing steps to solve the task. It would then choose the "best"…"not dissimilar" to how <a href="https://mastodon.social/tags/Google" class="mention hashtag" rel="nofollow noopener" target="_blank">#Google</a> <a href="https://mastodon.social/tags/AlphaGo" class="mention hashtag" rel="nofollow noopener" target="_blank">#AlphaGo</a> system…beat the world Go champion.

:rss: Hacker NewsChatGPT Learned to Reason [video] <a href="https://www.youtube.com/watch?v=PvDaPeQjxOE" rel="nofollow noopener" translate="no" target="_blank">https://www.youtube.com/watch?v=PvDaPeQjxOE</a> <a href="https://rss-mstdn.studiofreesia.com/tags/ycombinator" class="mention hashtag" rel="nofollow noopener" target="_blank">#ycombinator</a> <a href="https://rss-mstdn.studiofreesia.com/tags/AI_reasoning" class="mention hashtag" rel="nofollow noopener" target="_blank">#AI_reasoning</a> <a href="https://rss-mstdn.studiofreesia.com/tags/ChatGPT_explained" class="mention hashtag" rel="nofollow noopener" target="_blank">#ChatGPT_explained</a> <a href="https://rss-mstdn.studiofreesia.com/tags/artificial_intelligence" class="mention hashtag" rel="nofollow noopener" target="_blank">#artificial_intelligence</a> <a href="https://rss-mstdn.studiofreesia.com/tags/neural_networks" class="mention hashtag" rel="nofollow noopener" target="_blank">#neural_networks</a> <a href="https://rss-mstdn.studiofreesia.com/tags/Monte_Carlo_Tree_Search" class="mention hashtag" rel="nofollow noopener" target="_blank">#Monte_Carlo_Tree_Search</a> <a href="https://rss-mstdn.studiofreesia.com/tags/DeepMind" class="mention hashtag" rel="nofollow noopener" target="_blank">#DeepMind</a> <a href="https://rss-mstdn.studiofreesia.com/tags/AlphaGo" class="mention hashtag" rel="nofollow noopener" target="_blank">#AlphaGo</a> <a href="https://rss-mstdn.studiofreesia.com/tags/chess_AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#chess_AI</a> <a href="https://rss-mstdn.studiofreesia.com/tags/language_models" class="mention hashtag" rel="nofollow noopener" target="_blank">#language_models</a> <a href="https://rss-mstdn.studiofreesia.com/tags/machine_learning" class="mention hashtag" rel="nofollow noopener" target="_blank">#machine_learning</a> <a href="https://rss-mstdn.studiofreesia.com/tags/reinforcement_learning" class="mention hashtag" rel="nofollow noopener" target="_blank">#reinforcement_learning</a> <a href="https://rss-mstdn.studiofreesia.com/tags/deep_learning" class="mention hashtag" rel="nofollow noopener" target="_blank">#deep_learning</a> <a href="https://rss-mstdn.studiofreesia.com/tags/AI_history" class="mention hashtag" rel="nofollow noopener" target="_blank">#AI_history</a> <a href="https://rss-mstdn.studiofreesia.com/tags/GPT_training" class="mention hashtag" rel="nofollow noopener" target="_blank">#GPT_training</a> <a href="https://rss-mstdn.studiofreesia.com/tags/chain_of_thought" class="mention hashtag" rel="nofollow noopener" target="_blank">#chain_of_thought</a> <a href="https://rss-mstdn.studiofreesia.com/tags/AI_breakthrough" class="mention hashtag" rel="nofollow noopener" target="_blank">#AI_breakthrough</a> <a href="https://rss-mstdn.studiofreesia.com/tags/game_AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#game_AI</a> <a href="https://rss-mstdn.studiofreesia.com/tags/TD_Gammon" class="mention hashtag" rel="nofollow noopener" target="_blank">#TD_Gammon</a> <a href="https://rss-mstdn.studiofreesia.com/tags/MuZero" class="mention hashtag" rel="nofollow noopener" target="_blank">#MuZero</a> <a href="https://rss-mstdn.studiofreesia.com/tags/Claude_AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#Claude_AI</a> <a href="https://rss-mstdn.studiofreesia.com/tags/O1_AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#O1_AI</a> <a href="https://rss-mstdn.studiofreesia.com/tags/AI_algorithms" class="mention hashtag" rel="nofollow noopener" target="_blank">#AI_algorithms</a> <a href="https://rss-mstdn.studiofreesia.com/tags/AI_development" class="mention hashtag" rel="nofollow noopener" target="_blank">#AI_development</a> <a href="https://rss-mstdn.studiofreesia.com/tags/computer_reasoning" class="mention hashtag" rel="nofollow noopener" target="_blank">#computer_reasoning</a> <a href="https://rss-mstdn.studiofreesia.com/tags/AI_evolution" class="mention hashtag" rel="nofollow noopener" target="_blank">#AI_evolution</a> <a href="https://rss-mstdn.studiofreesia.com/tags/future_AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#future_AI</a>

🅴🆁🆄🅰 🇷🇺Странное чувство при осознании того, что видео-карта приобретена не только ради нормальной картинки в играх. Что эта вычислительная мощность активно используется и во время других игр, но где нужны обдуманные и взвешенные ходы, а не обсчёт задачек ради 3d-графики. Например, игра в #<a class="" href="https://hub.hubzilla.de/search?tag=%D0%93%D0%BE" rel="nofollow noopener" target="_blank">Го</a> — движки способные заменить людей выполняют неслабые такие нейросети в комбинации с #<a class="" href="https://hub.hubzilla.de/search?tag=MCTS" rel="nofollow noopener" target="_blank">MCTS</a> (Monte-Carlo Tree Search). Сперва это был #<a class="" href="https://hub.hubzilla.de/search?tag=LeelaZero" rel="nofollow noopener" target="_blank">LeelaZero</a>, являющийся повторением #<a class="" href="https://hub.hubzilla.de/search?tag=AlphaGo" rel="nofollow noopener" target="_blank">AlphaGo</a> Zero согласно его <a href="https://discovery.ucl.ac.uk/id/eprint/10045895/1/agz_unformatted_nature.pdf" rel="nofollow noopener" target="_blank">оригинальному описанию</a>. Теперь это #<a class="" href="https://hub.hubzilla.de/search?tag=KataGo" rel="nofollow noopener" target="_blank">KataGo</a>, в целом аналогичная, но с <a href="https://arxiv.org/abs/1902.10565" rel="nofollow noopener" target="_blank">рядом доработок</a> заточенных под игру #<a class="" href="https://hub.hubzilla.de/search?tag=%D0%93%D0%BE" rel="nofollow noopener" target="_blank">Го</a> и активно развиваемая/тренируемая, в то время как, работы над #<a class="" href="https://hub.hubzilla.de/search?tag=LeelaZero" rel="nofollow noopener" target="_blank">LeelaZero</a> прекратились в районе 2021 года. Очень может быть, что в скором времени, компьютерам нужна будет видеокарта как универсальный ускоритель общего назначения. Используемый и в локальных системах представления и анализа данных — самое элементарное создания динамических отчётов (dashboard'ов) заточенных под конкретного пользователя. Например, это могут быть различные «ассистенты», выполняющиеся на компьютере пользователя, к которым человек обращается с поручениями что-то узнать или выяснить. Ускорять специфичным «железом» (#<a class="" href="https://hub.hubzilla.de/search?tag=VLIW" rel="nofollow noopener" target="_blank">VLIW</a>, #<a class="" href="https://hub.hubzilla.de/search?tag=TensorFlow" rel="nofollow noopener" target="_blank">TensorFlow</a> ) надо будет не только парсинг/разбор запросов от человека на естественно языке (голосом, текстом), но и всю ту работу, которая должна быть проведена при создании запрошенного. #<a class="" href="https://hub.hubzilla.de/search?tag=AI" rel="nofollow noopener" target="_blank">AI</a> #<a class="" href="https://hub.hubzilla.de/search?tag=%D0%98%D0%98" rel="nofollow noopener" target="_blank">ИИ</a> #<a class="" href="https://hub.hubzilla.de/search?tag=games" rel="nofollow noopener" target="_blank">games</a> #<a class="" href="https://hub.hubzilla.de/search?tag=gaming" rel="nofollow noopener" target="_blank">gaming</a> #<a class="" href="https://hub.hubzilla.de/search?tag=%D0%B3%D0%BE" rel="nofollow noopener" target="_blank">го</a> #<a class="" href="https://hub.hubzilla.de/search?tag=igo" rel="nofollow noopener" target="_blank">igo</a> #<a class="" href="https://hub.hubzilla.de/search?tag=baduk" rel="nofollow noopener" target="_blank">baduk</a> #<a class="" href="https://hub.hubzilla.de/search?tag=%D0%B1%D0%B0%D0%B4%D1%83%D0%BA" rel="nofollow noopener" target="_blank">бадук</a> #<a class="" href="https://hub.hubzilla.de/search?tag=weiqi" rel="nofollow noopener" target="_blank">weiqi</a> #<a class="" href="https://hub.hubzilla.de/search?tag=%D0%B2%D1%8D%D0%B9%D1%86%D0%B8" rel="nofollow noopener" target="_blank">вэйци</a> #<a class="" href="https://hub.hubzilla.de/search?tag=OpenCL" rel="nofollow noopener" target="_blank">OpenCL</a> #<a class="" href="https://hub.hubzilla.de/search?tag=hardware" rel="nofollow noopener" target="_blank">hardware</a> #<a class="" href="https://hub.hubzilla.de/search?tag=lang_ru" rel="nofollow noopener" target="_blank">lang_ru</a>

🅴🆁🆄🅰 🇷🇺Попробовал как играют современные нейронки в #<a class="" href="https://hub.hubzilla.de/search?tag=%D0%93%D0%BE" rel="nofollow noopener" target="_blank">Го</a> на домашнем десктопе с простенькой видяхой, open source варианты, свободные. Затем, что современные значимые и серьёзные успехи «искусственного интеллекта» пошли в массы с эпопеи вокруг #<a class="" href="https://hub.hubzilla.de/search?tag=AlphaGo" rel="nofollow noopener" target="_blank">AlphaGo</a>, которое за три-четыре года развития изменилось сильно и в размерах и скорости работы, хорошо задокументировано и всячески изучено. Так вот, прошло изрядно лет уже с тех пор как AlphaGo остановилось в развитии, достигнув апогея (AlphaGo Zero), и где свободные аналоги? Пусть и заточенные именно для игры в Го и только для неё. Отыскался <a href="https://github.com/lightvector/KataGo?tab=readme-ov-file#" rel="nofollow noopener" target="_blank">движок KataGo</a>, вроде по тем же принципам, что и последние варианты AlphaGo, с уже обученными сетками, которые регулярно обновляются дообучаясь. #<a class="" href="https://hub.hubzilla.de/search?tag=KataGo" rel="nofollow noopener" target="_blank">KataGo</a> вариант использующий видеокарту, #<a class="" href="https://hub.hubzilla.de/search?tag=OpenCL" rel="nofollow noopener" target="_blank">OpenCL</a>, сходу не завёлся — пришлось погонять сперва вариант для #<a class="" href="https://hub.hubzilla.de/search?tag=CPU" rel="nofollow noopener" target="_blank">CPU</a>, чтобы подобрать GUI для использования движка: #<a class="" href="https://hub.hubzilla.de/search?tag=Sabaki" rel="nofollow noopener" target="_blank">Sabaki</a>, #<a class="" href="https://hub.hubzilla.de/search?tag=q5Go" rel="nofollow noopener" target="_blank">q5Go</a>. Заценив работу движка и шум системы охлаждения процессора — уже переключился на OpenCL-вариант. Для чего пришлось сносить из системы всё про #<a class="" href="https://hub.hubzilla.de/search?tag=Mesa" rel="nofollow noopener" target="_blank">Mesa</a> и ставить «opencl-amd» на #<a class="" href="https://hub.hubzilla.de/search?tag=ArchLinux" rel="nofollow noopener" target="_blank">ArchLinux</a>. И оно того стоит, не только потому что реально быстрее работает в плане ходов да подсчёта всякой аналитики, но главное комп перестал надрываться работой системы охлаждения. Памяти на видяхе KataGo отжирает порядка гигабайта, может полутора. Однако, у меня и режим работы выбран с дополнительной нейронкой для подражания человеку в плане манеры игры, ссылки на этот вариант работы движка есть в readme. Планшеты и мобильники? Для #<a class="" href="https://hub.hubzilla.de/search?tag=android" rel="nofollow noopener" target="_blank">android</a> тоже есть вариант KataGo — зовётся #<a class="" href="https://hub.hubzilla.de/search?tag=BadukAI" rel="nofollow noopener" target="_blank">BadukAI</a>, доступен и в Google'ом и Amazon'овском маркете <a href="https://play.google.com/store/apps/details?id=net.kir.baduk_ai&hl=en-US" rel="nofollow noopener" target="_blank">ссылка</a>, а так же через альтернативные клиенты. Если в #<a class="" href="https://hub.hubzilla.de/search?tag=Aurora" rel="nofollow noopener" target="_blank">Aurora</a> не работает вдруг поиск, то ссылку ту можно открыть/отправить в #<a class="" href="https://hub.hubzilla.de/search?tag=Aurora" rel="nofollow noopener" target="_blank">Aurora</a> и откроет спокойно. Что на счёт GUI? Прежде чем гонять движки разные, имеет смысл опробовать <a href="https://www.gnu.org/software/gnugo/" rel="nofollow noopener" target="_blank">GNU Go</a> — это который про игру в #<a class="" href="https://hub.hubzilla.de/search?tag=%D0%93%D0%BE" rel="nofollow noopener" target="_blank">Го</a> с компьютером, оно же «Бадук» у корейцев и «Вэйци» у китайцев (откуда якобы и пришло). GNU Go есть у всех #<a class="" href="https://hub.hubzilla.de/search?tag=linux" rel="nofollow noopener" target="_blank">linux</a> в репозиториях и доступен любому желающему поиграть в Го через #<a class="" href="https://hub.hubzilla.de/search?tag=Kigo" rel="nofollow noopener" target="_blank">Kigo</a>, #<a class="" href="https://hub.hubzilla.de/search?tag=qGo" rel="nofollow noopener" target="_blank">qGo</a>, #<a class="" href="https://hub.hubzilla.de/search?tag=q5Go" rel="nofollow noopener" target="_blank">q5Go</a>, #<a class="" href="https://hub.hubzilla.de/search?tag=Sabaki" rel="nofollow noopener" target="_blank">Sabaki</a>, #<a class="" href="https://hub.hubzilla.de/search?tag=KaTrain" rel="nofollow noopener" target="_blank">KaTrain</a>, #<a class="" href="https://hub.hubzilla.de/search?tag=Lizzie" rel="nofollow noopener" target="_blank">Lizzie</a>. Если выбранная GUI'шная софтина работает с GNU Go, то будет работать (должны) и со всякими другими движками для игры в Го, потому что используется gtp режим. Сложность лишь в том, что порой GNU Go путают с GNU'шным компилятором Golang — который называется иначе: <a href="https://go.dev/doc/install/gccgo" rel="nofollow noopener" target="_blank">GCC Go</a>. Альтернативы нейронкам? Есть вариант сугубо на базе #<a class="" href="https://hub.hubzilla.de/search?tag=MCTS" rel="nofollow noopener" target="_blank">MCTS</a> (который Monte Carlo tree search) — ощутимо получше GNU Go и более относительно современное — #<a class="" href="https://hub.hubzilla.de/search?tag=Pachi" rel="nofollow noopener" target="_blank">Pachi</a> Работает на десктопе #<a class="" href="https://hub.hubzilla.de/search?tag=Pachi" rel="nofollow noopener" target="_blank">Pachi</a> серьёзно нагружая систему, а так же имеет кучу заморочек по настройкам. Например, на тему подключения дополнительного движка ради нормальной игры в #<a class="" href="https://hub.hubzilla.de/search?tag=joseki" rel="nofollow noopener" target="_blank">joseki</a>, не сказать чтобы весёлый вариант. Однако, если ставить на android-устройство, то существует небольшой <a href="https://play.google.com/store/apps/details?id=net.lrstudios.android.pachi" rel="nofollow noopener" target="_blank">вариант</a>, размером менее трёх мегабайт, который вполне шустро работает. #<a class="" href="https://hub.hubzilla.de/search?tag=AI" rel="nofollow noopener" target="_blank">AI</a> #<a class="" href="https://hub.hubzilla.de/search?tag=%D0%98%D0%98" rel="nofollow noopener" target="_blank">ИИ</a> #<a class="" href="https://hub.hubzilla.de/search?tag=games" rel="nofollow noopener" target="_blank">games</a> #<a class="" href="https://hub.hubzilla.de/search?tag=gaming" rel="nofollow noopener" target="_blank">gaming</a> #<a class="" href="https://hub.hubzilla.de/search?tag=%D0%B3%D0%BE" rel="nofollow noopener" target="_blank">го</a> #<a class="" href="https://hub.hubzilla.de/search?tag=igo" rel="nofollow noopener" target="_blank">igo</a> #<a class="" href="https://hub.hubzilla.de/search?tag=baduk" rel="nofollow noopener" target="_blank">baduk</a> #<a class="" href="https://hub.hubzilla.de/search?tag=%D0%B1%D0%B0%D0%B4%D1%83%D0%BA" rel="nofollow noopener" target="_blank">бадук</a> #<a class="" href="https://hub.hubzilla.de/search?tag=weiqi" rel="nofollow noopener" target="_blank">weiqi</a> #<a class="" href="https://hub.hubzilla.de/search?tag=%D0%B2%D1%8D%D0%B9%D1%86%D0%B8" rel="nofollow noopener" target="_blank">вэйци</a> #<a class="" href="https://hub.hubzilla.de/search?tag=lang_ru" rel="nofollow noopener" target="_blank">lang_ru</a> @<a href="https://3zi.ru/@Russia" rel="nofollow noopener" target="_blank">Russia</a>

Matthias MProve<a href="https://recsys.social/@alansaid" class="u-url mention" rel="nofollow noopener" target="_blank">@alansaid</a> <a href="https://sigmoid.social/@Riedl" class="u-url mention" rel="nofollow noopener" target="_blank">@Riedl</a> à propos <a href="https://hci.social/tags/AlphaGo" class="mention hashtag" rel="nofollow noopener" target="_blank">#AlphaGo</a> >> <a href="https://hci.social/@mprove/111866463222208721" rel="nofollow noopener" translate="no" target="_blank">https://hci.social/@mprove/111866463222208721</a>

Arne BabenhauserheideWhen <a href="https://rollenspiel.social/tags/AlphaGo" class="mention hashtag" rel="nofollow noopener" target="_blank">#AlphaGo</a> cracked <a href="https://rollenspiel.social/tags/Go" class="mention hashtag" rel="nofollow noopener" target="_blank">#Go</a>, the holy grail of game <a href="https://rollenspiel.social/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#AI</a>, it proved that there are problems that we can currently only solve via a machine learning approach.Other approaches never managed more than mediocre play, AlphaGo beat the world class.Nine years later there are two types of AI:- Type 1 solves such problems. - Type 2 is <a href="https://rollenspiel.social/tags/bullshit" class="mention hashtag" rel="nofollow noopener" target="_blank">#bullshit</a>.<a href="https://www.draketo.de/zitate#alpha-go-nine-years-later" rel="nofollow noopener" translate="no" target="_blank">https://www.draketo.de/zitate#alpha-go-nine-years-later</a>

AltreconomiaScatole oscure o intelligenze aliene? Il caso del software AlphaGo e i fantasmi di Italo Calvino <a href="https://altreconomia.it/scatole-oscure-o-intelligenze-aliene-il-caso-del-software-alphago-e-i-fantasmi-di-italo-calvino/" rel="nofollow noopener" target="_blank">https://altreconomia.it/scatole-oscure-o-intelligenze-aliene-il-caso-del-software-alphago-e-i-fantasmi-di-italo-calvino/</a> <a href="https://sociale.network/tags/Intelligenzaartificiale" class="mention hashtag" rel="nofollow noopener" target="_blank">#Intelligenzaartificiale</a> <a href="https://sociale.network/tags/reinforcementlearning" class="mention hashtag" rel="nofollow noopener" target="_blank">#reinforcementlearning</a> <a href="https://sociale.network/tags/scatoleoscure" class="mention hashtag" rel="nofollow noopener" target="_blank">#scatoleoscure</a> <a href="https://sociale.network/tags/deeplearning" class="mention hashtag" rel="nofollow noopener" target="_blank">#deeplearning</a> <a href="https://sociale.network/tags/italocalvino" class="mention hashtag" rel="nofollow noopener" target="_blank">#italocalvino</a> <a href="https://sociale.network/tags/alanturing" class="mention hashtag" rel="nofollow noopener" target="_blank">#alanturing</a> <a href="https://sociale.network/tags/samaltman" class="mention hashtag" rel="nofollow noopener" target="_blank">#samaltman</a> <a href="https://sociale.network/tags/Opinioni" class="mention hashtag" rel="nofollow noopener" target="_blank">#Opinioni</a> <a href="https://sociale.network/tags/deepmind" class="mention hashtag" rel="nofollow noopener" target="_blank">#deepmind</a> <a href="https://sociale.network/tags/alphago" class="mention hashtag" rel="nofollow noopener" target="_blank">#alphago</a> <a href="https://sociale.network/tags/chatgpt" class="mention hashtag" rel="nofollow noopener" target="_blank">#chatgpt</a> <a href="https://sociale.network/tags/Harari" class="mention hashtag" rel="nofollow noopener" target="_blank">#Harari</a> <a href="https://sociale.network/tags/go" class="mention hashtag" rel="nofollow noopener" target="_blank">#go</a>

goldfishlaserI had long procrastinated listening to Lex Fridman's interview with David Silver about <a href="https://fosstodon.org/tags/alphago" class="mention hashtag" rel="nofollow noopener" target="_blank">#alphago</a> Linking below to my favorite part of the conversation, where he discusses self play and why there is no ceiling to improvements with increased computing power...<a href="https://youtu.be/uPUEq8d73JI?t=4512&si=GROx0rqS9YS3PAVO" rel="nofollow noopener" translate="no" target="_blank">https://youtu.be/uPUEq8d73JI?t=4512&si=GROx0rqS9YS3PAVO</a>

Michael Gisiger :mastodon:<a href="https://nerdculture.de/tags/OpenAI" class="mention hashtag" rel="nofollow noopener" target="_blank">#OpenAI</a> calls the <a href="https://nerdculture.de/tags/CoT" class="mention hashtag" rel="nofollow noopener" target="_blank">#CoT</a> <a href="https://nerdculture.de/tags/LLM" class="mention hashtag" rel="nofollow noopener" target="_blank">#LLM</a> like GPT-o1 ”reasoning models“. But are they really reasoning?"'Reasoning' is a semantic thing in my opinion," Kang [assistant professor in the computer science department at University of Illinois Urbana-Champaign] told The Register. "They are doing test-time scaling, which is roughly similar to what <a href="https://nerdculture.de/tags/AlphaGo" class="mention hashtag" rel="nofollow noopener" target="_blank">#AlphaGo</a> does. I don't know how to adjudicate semantic arguments, but I would anticipate that most people would consider this reasoning."<a href="https://www.theregister.com/2024/09/13/openai_rolls_out_reasoning_o1/" rel="nofollow noopener" translate="no" target="_blank">https://www.theregister.com/2024/09/13/openai_rolls_out_reasoning_o1/</a><a href="https://nerdculture.de/tags/ChatGPT" class="mention hashtag" rel="nofollow noopener" target="_blank">#ChatGPT</a> <a href="https://nerdculture.de/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#AI</a>

HabrСверхчеловеческий ИИ на основе открытых систем: Утопия или антиутопия?В погоне за созданием сверхчеловеческого интеллекта (ASI) исследователи ИИ все чаще обращаются к концепции открытых систем. Но что делает открытые системы столь многообещающими в контексте достижения ASI, и каковы фундаментальные принципы, лежащие в основе этого подхода? В этой статье мы погрузимся в техническое определение открытости через призму новизны и обучаемости, рассмотрим примеры существующих открытых систем, таких как AlphaGo, AdA и POET, и исследуем их потенциал на пути к ASI. Мы также обсудим ключевые подходы и проблемы безопасности и контроля, связанные с развитием открытых систем. Словом, полностью погрузимся в тему. Присоединяйтесь к нам в этом исследовании и рассуждению на тему open-source систем! Приятного прочтения:)<a href="https://habr.com/ru/companies/bothub/articles/835544/" rel="nofollow noopener" translate="no" target="_blank">https://habr.com/ru/companies/bothub/articles/835544/</a><a href="https://zhub.link/tags/%D0%B8%D1%81%D0%BA%D1%83%D1%81%D1%82%D0%B2%D0%B5%D0%BD%D0%BD%D1%8B%D0%B9_%D0%B8%D0%BD%D1%82%D0%B5%D0%BB%D0%BB%D0%B5%D0%BA%D1%82" class="mention hashtag" rel="nofollow noopener" target="_blank">#искуственный_интеллект</a> <a href="https://zhub.link/tags/%D0%BE%D1%82%D0%BA%D1%80%D1%8B%D1%82%D1%8B%D0%B5_%D1%81%D0%B8%D1%81%D1%82%D0%B5%D0%BC%D1%8B" class="mention hashtag" rel="nofollow noopener" target="_blank">#открытые_системы</a> <a href="https://zhub.link/tags/%D0%BE%D1%82%D0%BA%D1%80%D1%8B%D1%82%D1%8B%D0%B5_%D0%BC%D0%BE%D0%B4%D0%B5%D0%BB%D0%B8" class="mention hashtag" rel="nofollow noopener" target="_blank">#открытые_модели</a> <a href="https://zhub.link/tags/%D1%81%D0%B2%D0%B5%D1%80%D1%85%D1%87%D0%B5%D0%BB%D0%BE%D0%B2%D0%B5%D1%87%D0%B5%D1%81%D0%BA%D0%B8%D0%B9_%D0%B8%D0%B8" class="mention hashtag" rel="nofollow noopener" target="_blank">#сверхчеловеческий_ии</a> <a href="https://zhub.link/tags/%D0%BE%D0%B1%D1%83%D1%87%D0%B5%D0%BD%D0%B8%D0%B5_%D1%81_%D0%BF%D0%BE%D0%B4%D0%BA%D1%80%D0%B5%D0%BF%D0%BB%D0%B5%D0%BD%D0%B8%D0%B5%D0%BC" class="mention hashtag" rel="nofollow noopener" target="_blank">#обучение_с_подкреплением</a> <a href="https://zhub.link/tags/ada" class="mention hashtag" rel="nofollow noopener" target="_blank">#ada</a> <a href="https://zhub.link/tags/alphago" class="mention hashtag" rel="nofollow noopener" target="_blank">#alphago</a> <a href="https://zhub.link/tags/poetry" class="mention hashtag" rel="nofollow noopener" target="_blank">#poetry</a> <a href="https://zhub.link/tags/%D0%BC%D0%B0%D1%88%D0%B8%D0%BD%D0%BD%D0%BE%D0%B5_%D0%BE%D0%B1%D1%83%D1%87%D0%B5%D0%BD%D0%B8%D0%B5" class="mention hashtag" rel="nofollow noopener" target="_blank">#машинное_обучение</a>

IvesI used to play a lot of <a href="https://mstdn.social/tags/Go" class="mention hashtag" rel="nofollow noopener" target="_blank">#Go</a> (poorly). It was fun to play a game where humans still beat computers. Until 2015 when <a href="https://mstdn.social/tags/AlphaGo" class="mention hashtag" rel="nofollow noopener" target="_blank">#AlphaGo</a> beat <a href="https://mstdn.social/tags/LeeSedol" class="mention hashtag" rel="nofollow noopener" target="_blank">#LeeSedol</a>, a 9-dan professional. It felt sad, to have that taken away from humanity. But a year ago someone found a strategy that works against a range of Go playing programs, despite them being a lot stronger nowadays. Ironically it's quite simple and would not work against a human. So humans beat computers again in Go (<a href="https://mstdn.social/tags/WeiQi" class="mention hashtag" rel="nofollow noopener" target="_blank">#WeiQi</a>, <a href="https://mstdn.social/tags/Baduk" class="mention hashtag" rel="nofollow noopener" target="_blank">#Baduk</a>). Yay!<a href="https://arstechnica.com/ai/2024/07/superhuman-go-ais-still-have-trouble-defending-against-these-simple-exploits/" rel="nofollow noopener" translate="no" target="_blank">https://arstechnica.com/ai/2024/07/superhuman-go-ais-still-have-trouble-defending-against-these-simple-exploits/</a>

Princess (xe/xem)I was quite impressed when I found out AI was able to beat the best human Go player. Fast forward a few years and humans are able to win from computers again by exploiting a weakness in AI. I think this says everything about the current state of AI: it looks impressive, but is so easy to trick. <a href="https://www.capstan.be/first-human-player-since-2016-to-defeats-ai-at-board-game-go-does-so-by-exploiting-a-weakness-in-the-system/" rel="nofollow noopener" translate="no" target="_blank">https://www.capstan.be/first-human-player-since-2016-to-defeats-ai-at-board-game-go-does-so-by-exploiting-a-weakness-in-the-system/</a> <a href="https://mas.to/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#AI</a> <a href="https://mas.to/tags/ArtificialIntelligence" class="mention hashtag" rel="nofollow noopener" target="_blank">#ArtificialIntelligence</a> <a href="https://mas.to/tags/AlphaGo" class="mention hashtag" rel="nofollow noopener" target="_blank">#AlphaGo</a> <a href="https://mas.to/tags/trick" class="mention hashtag" rel="nofollow noopener" target="_blank">#trick</a>

Recent searches

Search options

Administered by:

Server stats:

#alphago