<?xml version="1.0" encoding="UTF-8"?><oembed><type>video</type><version>1.0</version><html>&lt;iframe src=&quot;https://www.loom.com/embed/8cdd70ba5d6348abb5065f293fbdb7d1&quot; frameborder=&quot;0&quot; width=&quot;1920&quot; height=&quot;1440&quot; webkitallowfullscreen mozallowfullscreen allowfullscreen&gt;&lt;/iframe&gt;</html><height>1440</height><width>1920</width><provider_name>Loom</provider_name><provider_url>https://www.loom.com</provider_url><thumbnail_height>1440</thumbnail_height><thumbnail_width>1920</thumbnail_width><thumbnail_url>https://cdn.loom.com/sessions/thumbnails/8cdd70ba5d6348abb5065f293fbdb7d1-2ca9421ff4e33638.gif</thumbnail_url><duration>233.248</duration><title>World Modeling AI Bot in Pokémon Battles</title><description>This Loom demonstrates a Pokemon Battling project to explain why the state space is challenging and how an AI bot can handle it using world modeling. The presenter shows a custom Pikachu with its own item, moves like Agility, Body Slam, Iron Tail, and Tundra, and stat modifications for a fast, attacking build. They then battle a Latios, trying different move strategies but initially getting beaten due to speed. The AI bot’s internal reasoning is displayed, including its input to the world model and its predicted next state and best action, such as switching Pokemon.</description></oembed>