<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Machine-Learning on Karpoke - Just Another Blog</title><link>http://karpoke.ignaciocano.com/tags/machine-learning/</link><description>Recent content in Machine-Learning on Karpoke - Just Another Blog</description><generator>Hugo -- 0.167.0</generator><language>es</language><lastBuildDate>Wed, 29 Apr 2026 12:00:00 +0100</lastBuildDate><atom:link href="http://karpoke.ignaciocano.com/tags/machine-learning/index.xml" rel="self" type="application/rss+xml"/><item><title>This AI Was Trained Only on Pre-1930 Text. We Asked It About Hitler, Stocks, and the Future - Decrypt</title><link>http://karpoke.ignaciocano.com/2026/04/29/this-ai-was-trained-only-on-pre-1930-text-we-asked-it-about-hitler-stocks-and-th/</link><pubDate>Wed, 29 Apr 2026 12:00:00 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2026/04/29/this-ai-was-trained-only-on-pre-1930-text-we-asked-it-about-hitler-stocks-and-th/</guid><description>&lt;blockquote&gt;
&lt;p&gt;Talkie-1930 is an AI model that has never heard of the internet, World
War II, or modern politics. The results are fascinating.&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» Jose Antonio Lanz | &lt;a href="https://decrypt.co/366015/talkie-1930-ai-model-pre-1931-training-hitler-stocks"&gt;decrypt.co&lt;/a&gt;&lt;/p&gt;</description></item><item><title>200 Blog Posts To Learn About Artificial Intelligence Trends | HackerNoon</title><link>http://karpoke.ignaciocano.com/2026/04/26/200-blog-posts-to-learn-about-artificial-intelligence-trends-hackernoon/</link><pubDate>Sun, 26 Apr 2026 12:00:00 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2026/04/26/200-blog-posts-to-learn-about-artificial-intelligence-trends-hackernoon/</guid><description>&lt;blockquote&gt;
&lt;p&gt;Learn everything you need to know about Artificial Intelligence Trends
via these 200 free HackerNoon blog posts.&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» hackernoon.com | &lt;a href="https://hackernoon.com/200-blog-posts-to-learn-about-artificial-intelligence-trends"&gt;hackernoon.com&lt;/a&gt;&lt;/p&gt;</description></item><item><title>480 Blog Posts To Learn About Ai Applications | HackerNoon</title><link>http://karpoke.ignaciocano.com/2026/04/20/480-blog-posts-to-learn-about-ai-applications-hackernoon/</link><pubDate>Mon, 20 Apr 2026 12:00:00 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2026/04/20/480-blog-posts-to-learn-about-ai-applications-hackernoon/</guid><description>&lt;blockquote&gt;
&lt;p&gt;Learn everything you need to know about Ai Applications via these 480
free HackerNoon blog posts.&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» hackernoon.com | &lt;a href="https://hackernoon.com/480-blog-posts-to-learn-about-ai-applications"&gt;hackernoon.com&lt;/a&gt;&lt;/p&gt;</description></item><item><title>Fundamentos de Inteligencia Artificial - Tercera parte: Glosario</title><link>http://karpoke.ignaciocano.com/2026/03/13/fundamentos-de-inteligencia-artificial-tercera-parte-glosario/</link><pubDate>Fri, 13 Mar 2026 12:00:00 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2026/03/13/fundamentos-de-inteligencia-artificial-tercera-parte-glosario/</guid><description>&lt;blockquote&gt;
&lt;p&gt;La Inteligencia Artificial tiene mucha relación con el Linuxverso, por
ello, hoy te ofrecemos esta tercera parte para aprender más sobre ella.&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» Jose Albert | &lt;a href="https://blog.desdelinux.net/inteligencia-artificial-ia-para-linuxeros-tercera-parte/"&gt;blog.desdelinux.net&lt;/a&gt;&lt;/p&gt;</description></item><item><title>AI Wrapped: The 14 AI terms you couldn’t avoid in 2025</title><link>http://karpoke.ignaciocano.com/2025/12/25/ai-wrapped-the-14-ai-terms-you-couldnt-avoid-in-2025/</link><pubDate>Thu, 25 Dec 2025 12:15:04 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2025/12/25/ai-wrapped-the-14-ai-terms-you-couldnt-avoid-in-2025/</guid><description>&lt;blockquote&gt;
&lt;p&gt;From “superintelligence” to “slop,” here are the words and phrases that
defined another year of AI craziness.&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» Caiwei Chen,Will Douglas Heaven,Michelle Kim,James O&amp;rsquo;Donnell,Rhiannon Williams | &lt;a href="https://www.technologyreview.com/2025/12/25/1130298/ai-wrapped-the-14-ai-terms-you-couldnt-avoid-in-2025/"&gt;technologyreview.com&lt;/a&gt;&lt;/p&gt;</description></item><item><title>60 of our biggest AI announcements in 2025</title><link>http://karpoke.ignaciocano.com/2025/12/23/60-of-our-biggest-ai-announcements-in-2025/</link><pubDate>Tue, 23 Dec 2025 07:30:48 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2025/12/23/60-of-our-biggest-ai-announcements-in-2025/</guid><description>&lt;blockquote&gt;
&lt;p&gt;Look back on Google AI news in 2025 across Gemini, Search, Pixel and
more products.&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» Keyword Team | &lt;a href="https://blog.google/technology/ai/google-ai-news-recap-2025/"&gt;blog.google&lt;/a&gt;&lt;/p&gt;</description></item><item><title>El enigma de la IA generativa: hace cosas asombrosas, pero no sabemos bien por qué</title><link>http://karpoke.ignaciocano.com/2024/03/19/el-enigma-de-la-ia-generativa-hace-cosas-asombrosas-pero-no-sabemos-bien-por-que/</link><pubDate>Tue, 19 Mar 2024 10:55:35 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2024/03/19/el-enigma-de-la-ia-generativa-hace-cosas-asombrosas-pero-no-sabemos-bien-por-que/</guid><description>&lt;blockquote&gt;
&lt;p&gt;Somos la revista sobre tecnología más antigua del mundo y la autoridad
global en el futuro de la tecnología en internet.&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» MIT Technology Review en español - Somos la revista sobre tecnología más antigua del mundo y la autoridad global en el futuro de la tecnología en internet. | &lt;a href="https://www.technologyreview.es/s/16235/el-enigma-de-la-ia-generativa-hace-cosas-asombrosas-pero-no-sabemos-bien-por-que"&gt;technologyreview.es&lt;/a&gt;&lt;/p&gt;</description></item><item><title>The Noonification: Turn GPT-4 Into Your Expert: Fine-Tuning Large Language Models Easily (9/20/2023)</title><link>http://karpoke.ignaciocano.com/2023/09/20/the-noonification-turn-gpt-4-into-your-expert-fine-tuning-large-language-models/</link><pubDate>Wed, 20 Sep 2023 23:44:30 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2023/09/20/the-noonification-turn-gpt-4-into-your-expert-fine-tuning-large-language-models/</guid><description>&lt;blockquote&gt;
&lt;p&gt;The Noonification: Turn GPT-4 Into Your Expert: Fine-Tuning Large Language Models Easily (9/20/2023)&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» hackernoon.com | &lt;a href="https://hackernoon.com/9-20-2023-noonification"&gt;hackernoon.com&lt;/a&gt;&lt;/p&gt;</description></item><item><title>Basic Terminologies Large Language Models</title><link>http://karpoke.ignaciocano.com/2023/08/21/basic-terminologies-large-language-models/</link><pubDate>Mon, 21 Aug 2023 08:34:59 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2023/08/21/basic-terminologies-large-language-models/</guid><description>&lt;blockquote&gt;
&lt;p&gt;Basic Terminologies Large Language Models&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» frankdenneman.nl | &lt;a href="https://frankdenneman.nl/2023/08/18/basic-terminologies-large-language-models/"&gt;frankdenneman.nl&lt;/a&gt;&lt;/p&gt;</description></item><item><title>What is the future of AI?</title><link>http://karpoke.ignaciocano.com/2023/03/20/what-is-the-future-of-ai/</link><pubDate>Mon, 20 Mar 2023 10:46:12 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2023/03/20/what-is-the-future-of-ai/</guid><description>&lt;blockquote&gt;
&lt;p&gt;Recently, when I saw another article about ChatGPT, I thought, &amp;ldquo;What&amp;rsquo;s
the big deal of it? Why is&amp;hellip;&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» DEV Community | &lt;a href="https://dev.to/brownrita460/what-is-the-future-of-ai-n4o"&gt;dev.to&lt;/a&gt;&lt;/p&gt;</description></item><item><title>Open AI | Una mirada a la nuevas capacidades de GPT-4 [EN]</title><link>http://karpoke.ignaciocano.com/2023/03/19/open-ai-una-mirada-a-la-nuevas-capacidades-de-gpt-4-en/</link><pubDate>Sun, 19 Mar 2023 21:42:05 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2023/03/19/open-ai-una-mirada-a-la-nuevas-capacidades-de-gpt-4-en/</guid><description>&lt;blockquote&gt;
&lt;p&gt;Open AI | Una mirada a la nuevas capacidades de GPT-4 [EN]&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» openai.com | &lt;a href="https://openai.com/research/gpt-4"&gt;openai.com&lt;/a&gt;&lt;/p&gt;</description></item><item><title>GPT-4 de OpenAI – Primeras impresiones / Dot CSV</title><link>http://karpoke.ignaciocano.com/2023/03/17/gpt-4-de-openai-primeras-impresiones-dot-csv/</link><pubDate>Fri, 17 Mar 2023 12:04:28 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2023/03/17/gpt-4-de-openai-primeras-impresiones-dot-csv/</guid><description>&lt;blockquote&gt;
&lt;p&gt;GPT-4 de OpenAI – Primeras impresiones / Dot CSV&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» PROYECTO TIC TAC | &lt;a href="https://proyectotictac.com/2023/03/17/gpt-4-openai-primeras-impresiones-dot-csv/"&gt;proyectotictac.com&lt;/a&gt;&lt;/p&gt;</description></item><item><title>Noam Chomsky: La falsa promesa del ChatGPT</title><link>http://karpoke.ignaciocano.com/2023/03/14/noam-chomsky-la-falsa-promesa-del-chatgpt/</link><pubDate>Tue, 14 Mar 2023 15:32:10 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2023/03/14/noam-chomsky-la-falsa-promesa-del-chatgpt/</guid><description>&lt;blockquote&gt;
&lt;p&gt;Noam Chomsky: La falsa promesa del ChatGPT&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» hypernoir.com | &lt;a href="https://hypernoir.com/noam-chomsky-la-falsa-promesa-del-chatgpt/"&gt;hypernoir.com&lt;/a&gt;&lt;/p&gt;</description></item><item><title>The inside story of how ChatGPT was built from the people who made it</title><link>http://karpoke.ignaciocano.com/2023/03/05/the-inside-story-of-how-chatgpt-was-built-from-the-people-who-made-it/</link><pubDate>Sun, 05 Mar 2023 21:56:43 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2023/03/05/the-inside-story-of-how-chatgpt-was-built-from-the-people-who-made-it/</guid><description>&lt;blockquote&gt;
&lt;p&gt;Exclusive conversations that take us behind the scenes of a cultural
phenomenon.&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» Will Douglas Heaven | &lt;a href="https://www.technologyreview.com/2023/03/03/1069311/inside-story-oral-history-how-chatgpt-built-openai/"&gt;technologyreview.com&lt;/a&gt;&lt;/p&gt;</description></item><item><title>The Difference Between ChatGPT and GPT-3</title><link>http://karpoke.ignaciocano.com/2022/12/12/the-difference-between-chatgpt-and-gpt-3/</link><pubDate>Mon, 12 Dec 2022 20:48:48 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2022/12/12/the-difference-between-chatgpt-and-gpt-3/</guid><description>&lt;blockquote&gt;
&lt;p&gt;ChatGPT was announced a few days ago, and it has been a big topic of
discussion in the tech world,&amp;hellip;&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» DEV Community | &lt;a href="https://dev.to/ben/the-difference-between-chatgpt-and-gpt-3-19dh"&gt;dev.to&lt;/a&gt;&lt;/p&gt;</description></item><item><title>Stable Diffusion 2.0, una AI capaz de sintetizar y modificar imágenes</title><link>http://karpoke.ignaciocano.com/2022/11/27/stable-diffusion-2-0-una-ai-capaz-de-sintetizar-y-modificar-imagenes/</link><pubDate>Sun, 27 Nov 2022 11:20:20 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2022/11/27/stable-diffusion-2-0-una-ai-capaz-de-sintetizar-y-modificar-imagenes/</guid><description>&lt;blockquote&gt;
&lt;p&gt;La nueva versión de Stable Diffusion 2.0 rediseña los componentes clave
del modelo y mejora ciertas características como la ampliación&amp;hellip;&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» David Y. Naranjo | &lt;a href="https://blog.desdelinux.net/stable-diffusion-2-0-una-ai-capaz-de-sintetizar-y-modificar-imagenes/"&gt;blog.desdelinux.net&lt;/a&gt;&lt;/p&gt;</description></item><item><title>The Best AI Articles of October 2022</title><link>http://karpoke.ignaciocano.com/2022/11/03/the-best-ai-articles-of-october-2022/</link><pubDate>Thu, 03 Nov 2022 01:05:39 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2022/11/03/the-best-ai-articles-of-october-2022/</guid><description>&lt;blockquote&gt;
&lt;p&gt;Support vector machines, decision trees, and AI-generated content are
some of the topics in the best AI articles of October.&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» hackernoon.com | &lt;a href="https://hackernoon.com/the-best-ai-articles-of-october-2022"&gt;hackernoon.com&lt;/a&gt;&lt;/p&gt;</description></item><item><title>Crean un traductor que descifra las emociones de los cerdos a partir de sus gruñidos</title><link>http://karpoke.ignaciocano.com/2022/03/13/crean-un-traductor-que-descifra-las-emociones-de-los-cerdos-a-partir-de-sus-grun/</link><pubDate>Sun, 13 Mar 2022 23:20:38 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2022/03/13/crean-un-traductor-que-descifra-las-emociones-de-los-cerdos-a-partir-de-sus-grun/</guid><description>&lt;blockquote&gt;
&lt;p&gt;Es común que al darse interacciones entre personas que hablan un idioma
diferente, la comunicación entre ambas resulte en una experiencia
incomoda y nada satisfactoria si ninguno conoce cómo expresarse en el
idioma del otro. Es por ello que muchos deciden romper las barreras…&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» Alexander Alvarado | &lt;a href="https://wwwhatsnew.com/2022/03/11/crean-un-traductor-que-descifra-las-emociones-de-los-cerdos-a-partir-de-sus-grunidos/"&gt;wwwhatsnew.com&lt;/a&gt;&lt;/p&gt;</description></item><item><title>An AI That Can Resurrect Memories</title><link>http://karpoke.ignaciocano.com/2022/03/12/an-ai-that-can-resurrect-memories/</link><pubDate>Sat, 12 Mar 2022 10:35:31 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2022/03/12/an-ai-that-can-resurrect-memories/</guid><description>&lt;blockquote&gt;
&lt;p&gt;Do you also have old pictures of yourself or close ones that didn’t age
well or that you, or your parents, took before we could produce
high-quality images? I do, and I felt like those memories were damaged
forever. Boy, was I wrong!&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» hackernoon.com | &lt;a href="https://hackernoon.com/an-ai-that-can-resurrect-memories"&gt;hackernoon.com&lt;/a&gt;&lt;/p&gt;</description></item><item><title>How to Train Self-Parking Car Using Genetic Algorithm (&lt;500 Lines of Code)</title><link>http://karpoke.ignaciocano.com/2021/10/15/how-to-train-self-parking-car-using-genetic-algorithm-500-lines-of-code/</link><pubDate>Fri, 15 Oct 2021 08:29:38 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2021/10/15/how-to-train-self-parking-car-using-genetic-algorithm-500-lines-of-code/</guid><description>&lt;blockquote&gt;
&lt;p&gt;Training the car to do self-parking using a genetic algorithm&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» trekhleb.dev | &lt;a href="https://trekhleb.dev/blog/2021/self-parking-car-evolution/"&gt;trekhleb.dev&lt;/a&gt;&lt;/p&gt;</description></item><item><title>Corporate AI Transformation: How to Not Throw Mud Against the Wall</title><link>http://karpoke.ignaciocano.com/2021/09/18/corporate-ai-transformation-how-to-not-throw-mud-against-the-wall/</link><pubDate>Sat, 18 Sep 2021 17:34:33 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2021/09/18/corporate-ai-transformation-how-to-not-throw-mud-against-the-wall/</guid><description>&lt;blockquote&gt;
&lt;p&gt;In order for existing organizations to undergo a successful AI
transformation, companies must remember that they are creating
human-machine teams.&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» hackernoon.com | &lt;a href="https://hackernoon.com/corporate-ai-transformation-how-to-not-throw-mud-against-the-wall"&gt;hackernoon.com&lt;/a&gt;&lt;/p&gt;</description></item><item><title>TensorFlow 2 'hello world'</title><link>http://karpoke.ignaciocano.com/2021/02/26/tensorflow-2-hello-world/</link><pubDate>Fri, 26 Feb 2021 22:11:40 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2021/02/26/tensorflow-2-hello-world/</guid><description>&lt;blockquote&gt;
&lt;p&gt;A TensorFlow 2 program that learns the cost per gram of chocolate by
training a linear model on example chocolate bar weights and prices.&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» jameshfisher.com | &lt;a href="https://jameshfisher.com/2021/02/17/tensorflow-2-hello-world/"&gt;jameshfisher.com&lt;/a&gt;&lt;/p&gt;</description></item><item><title>GPT3 explicado</title><link>http://karpoke.ignaciocano.com/2020/12/06/gpt3-explicado/</link><pubDate>Sun, 06 Dec 2020 11:02:42 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2020/12/06/gpt3-explicado/</guid><description>&lt;blockquote&gt;
&lt;p&gt;GPT3 explicado&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» planaspa.com | &lt;a href="https://planaspa.com/2020/07/26/GPT-3.html"&gt;planaspa.com&lt;/a&gt;&lt;/p&gt;</description></item><item><title>Basics Of Natural Language Processing in 10 Minutes</title><link>http://karpoke.ignaciocano.com/2020/11/28/basics-of-natural-language-processing-in-10-minutes/</link><pubDate>Sat, 28 Nov 2020 23:27:04 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2020/11/28/basics-of-natural-language-processing-in-10-minutes/</guid><description>&lt;blockquote&gt;
&lt;p&gt;Hello, there You are here because you also want to learn natural
language processing as quickly as po&amp;hellip;&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» DEV Community | &lt;a href="https://dev.to/abhayparashar31/basics-of-natural-language-processing-in-10-minutes-5fmg"&gt;dev.to&lt;/a&gt;&lt;/p&gt;</description></item><item><title>Alguien dejó suelto un bot GPT-3 en Reddit. Y no acabó bien (ING).</title><link>http://karpoke.ignaciocano.com/2020/10/08/alguien-dejo-suelto-un-bot-gpt-3-en-reddit-y-no-acabo-bien-ing/</link><pubDate>Thu, 08 Oct 2020 10:16:29 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2020/10/08/alguien-dejo-suelto-un-bot-gpt-3-en-reddit-y-no-acabo-bien-ing/</guid><description>&lt;blockquote&gt;
&lt;p&gt;A GPT-3-powered bot has been caught posing as a human on Reddit after
more than a week of rampant posting about some seriously sensitive
subjects.&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» Thomas Macaulay | &lt;a href="https://thenextweb.com/neural/2020/10/07/someone-let-a-gpt-3-bot-loose-on-reddit-it-didnt-end-well/amp/"&gt;thenextweb.com&lt;/a&gt;&lt;/p&gt;</description></item><item><title>El enorme blockhead que es GPT-3</title><link>http://karpoke.ignaciocano.com/2020/09/06/el-enorme-blockhead-que-es-gpt-3/</link><pubDate>Sun, 06 Sep 2020 09:47:49 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2020/09/06/el-enorme-blockhead-que-es-gpt-3/</guid><description>&lt;blockquote&gt;
&lt;p&gt;Cuando pensamos en cómo implementar un programa capaz de manejar
competentemente un lenguaje, lo primero que se nos pasa por la cabeza es
enseñar a la máquina gramática. Hay que saber de sujetos, p…&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» La Máquina de Von Neumann | &lt;a href="https://vonneumannmachine.wordpress.com/2020/09/05/el-enorme-blokhead-que-es-gpt-3/"&gt;vonneumannmachine.wordpress.com&lt;/a&gt;&lt;/p&gt;</description></item><item><title>Best Tensorflow Interview Questions</title><link>http://karpoke.ignaciocano.com/2020/08/29/best-tensorflow-interview-questions/</link><pubDate>Sat, 29 Aug 2020 10:15:32 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2020/08/29/best-tensorflow-interview-questions/</guid><description>&lt;blockquote&gt;
&lt;p&gt;Prepare for your TensorFlow interview - Explore the best and most
relevant interview questions to ace your technical assessment.&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» Partner Content | &lt;a href="https://fossbytes.com/best-tensorflow-interview-questions/"&gt;fossbytes.com&lt;/a&gt;&lt;/p&gt;</description></item><item><title>GPT-3: The Gigantic Deep Learning Model Created by OpenAI</title><link>http://karpoke.ignaciocano.com/2020/08/03/gpt-3-the-gigantic-deep-learning-model-created-by-openai/</link><pubDate>Mon, 03 Aug 2020 17:27:55 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2020/08/03/gpt-3-the-gigantic-deep-learning-model-created-by-openai/</guid><description>&lt;blockquote&gt;
&lt;p&gt;GPT-3: The Gigantic Deep Learning Model Created by OpenAI&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» hackernoon.com | &lt;a href="https://hackernoon.com/gpt-3-the-gigantic-deep-learning-model-created-by-openai-t0143u60"&gt;hackernoon.com&lt;/a&gt;&lt;/p&gt;</description></item><item><title>Top 10 best uses of GPT-3</title><link>http://karpoke.ignaciocano.com/2020/07/27/top-10-best-uses-of-gpt-3/</link><pubDate>Mon, 27 Jul 2020 15:54:12 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2020/07/27/top-10-best-uses-of-gpt-3/</guid><description>&lt;blockquote&gt;
&lt;p&gt;Hi everyone, After watching a lot of buzz about GPT-3 on social media
and how it is about the change&amp;hellip;&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» DEV Community | &lt;a href="https://dev.to/amananandrai/top-10-best-uses-of-gpt-3-175n"&gt;dev.to&lt;/a&gt;&lt;/p&gt;</description></item><item><title>GPT-What: Intro to the next big thing</title><link>http://karpoke.ignaciocano.com/2020/07/27/gpt-what-intro-to-the-next-big-thing/</link><pubDate>Mon, 27 Jul 2020 15:54:00 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2020/07/27/gpt-what-intro-to-the-next-big-thing/</guid><description>&lt;blockquote&gt;
&lt;p&gt;Breaking down OpenAI’s much-hyped language model GPT-3 and looking at
its promises and shortcomings&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» DEV Community | &lt;a href="https://dev.to/jesalg/gpt-what-intro-to-the-next-big-thing-1jc1"&gt;dev.to&lt;/a&gt;&lt;/p&gt;</description></item><item><title>Codementor: K-Nearest Neighbors explained</title><link>http://karpoke.ignaciocano.com/2020/02/12/codementor-k-nearest-neighbors-explained/</link><pubDate>Wed, 12 Feb 2020 13:34:00 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2020/02/12/codementor-k-nearest-neighbors-explained/</guid><description>&lt;blockquote&gt;
&lt;p&gt;In this post, I explain the intuition and logic behind KNN algorithm and
show simple implementation written in pure pandas which yield 98%
accuracy on the IRIS dataset.&lt;/p&gt;
&lt;/blockquote&gt;
&lt;p&gt;» Volodymyr Lut | &lt;a href="https://www.codementor.io/@volodymyrlut/k-nearest-neighbors-explained-13jm2ii1gx"&gt;codementor.io&lt;/a&gt;&lt;/p&gt;</description></item><item><title>How We Teach Computers to Understand Pictures</title><link>http://karpoke.ignaciocano.com/2016/06/29/how-we-teach-computers-to-understand-pictures/</link><pubDate>Wed, 29 Jun 2016 01:33:00 +0100</pubDate><guid>http://karpoke.ignaciocano.com/2016/06/29/how-we-teach-computers-to-understand-pictures/</guid><description>&lt;blockquote&gt;
&lt;p&gt;When a very young child looks at a picture, she can identify simple elements:
”cat,” ”book,” ”chair.” Now, computers are getting smart enough to do that
too. What’s next? In a thrilling talk, computer vision expert Fei-Fei Li
describes the state of the art — including the database of 15 million photos
her team built to ”teach” a computer to understand pictures — and the key
insights yet to come.&lt;/p&gt;
&lt;/blockquote&gt;</description></item></channel></rss>