Memory Meets Psychology - Claude Plays Pokemon: How It works, How to improve it

Subscribers:
6,310
Published on ● Video Link: https://www.youtube.com/watch?v=1Yaf6OSCRkk



Duration: 0:00
1,578 views
34


Inspired by Claude Plays Pokemon (CPP) ( https://www.twitch.tv/claudeplayspokemon), I will be implementing AgentJo Plays Pokemon with a few friends (advance thanks to Dylan Chia Tian for the livestream interface)! It will be open-sourced, and will be used to test out multimodal memory, goal-setting, adaptive memory updates and planning. See the github repo here: https://github.com/tanchongmin/agentjo/tree/main/projects/Pokemon

The first version of the CPP run saw it not being able to remember things, exploring the same things again with enthusiasm, and stuck in a loop in Mt. Moon.

The second version used context-based memory files to store memory for different contexts, and fares better. However, it still forgets important information.

I posit that human psychology can help in improving memory and adaptation.

We have emotions that can help to priorities memory encoding and retrieval - fear memories are stored more strongly, boring memories can be easily forgotten.

We also have the event boundary of the hippocampus to better store context when transiting between contexts. This is a useful thing to have in AI, so we can have memory that is more context-based when encoding.

Video:    • Memory Meets Psychology - Claude Plays Pok...  

~~

Slides: https://github.com/tanchongmin/agentjo/blob/main/paper_reviews/Claude Plays Pokemon Insights.pdf

~~

0:00 Introduction to Claude Plays Pokemon
1:34 AgentJo Plays Pokemon
3:51 Overview of Claude Plays Pokemon
23:16 Memory Issues
32:52 Psychology Insights for better memory retrieval and encoding
50:11 Image-based memory
58:25 Algorithms for navigation
1:03:57 Discussion
1:28:11 AgentJo Logo Design Competition

~~

AI and ML enthusiast. Likes to think about the essences behind breakthroughs of AI and explain it in a simple and relatable way. Also, I am an avid game creator.

Discord: https://discord.gg/bzp87AHJy5
LinkedIn: https://www.linkedin.com/in/chong-min-tan-94652288/
Online AI blog: https://delvingintotech.wordpress.com/
Twitter: https://twitter.com/johntanchongmin
Try out my games here: https://simmer.io/@chongmin




Other Videos By John Tan Chong Min


2025-08-04Reasoning without Language - Deep Dive into 27 mil parameter Hierarchical Reasoning Model
2025-07-28No need for symbolic programs for Math? Natural language approach to IMO
2025-07-21How many instructions can LLMs follow at once?
2025-07-15Arjo Chakravarty: Indoor Localisation with Visual Language Models (VLMs)
2025-07-14MemOS: A Paradigm Shift to Memory as a First Class Citizen for LLMs
2025-07-07Multimodal Query for Images: Text/Image Multimodal Query with Negative Filter and Folder Selection
2025-06-30Universal Filter (Part 4 - Finale): Knowledge/Memory, Reflection, Communication between Individuals
2025-06-23Universal Filter (Part 3): Learning the Filters, Universal Database, Individual Knowledge Base
2025-06-16Universal Filter (Part 2): Time, Akashic Records, Individual Mind-based, Body-based memory
2025-06-04Good Vibes Only with Dylan Chia: Lyria (Music), Veo3 (Video), Gamma (Slides), GitHub Copilot (Code)
2025-03-10Memory Meets Psychology - Claude Plays Pokemon: How It works, How to improve it
2025-02-24Vibe Coding: How to use LLM prompts to code effectively!
2025-01-26PhD Thesis Overview (Part 2): LLMs for ARC-AGI, Task-Based Memory-Infused Learning, Plan for AgentJo
2025-01-20PhD Thesis Overview (Part 1): Reward is not enough; Towards Goal-Directed, Memory-based Learning
2024-12-04AgentJo CV Generator: Generate your CV by searching for your profile on the web!
2024-11-11Can LLMs be used in self-driving? CoMAL: Collaborative Multi-Agent LLM for Mixed Autonomy Traffic
2024-10-28From TaskGen to AgentJo: Creating My Life Dream of Fast Learning and Adaptable Agents
2024-10-21Tian Yu X John: Discussing Practical Gen AI Tips for Image Prompting
2024-10-08Jiafei Duan: Uncovering the 'Right' Representations for Multimodal LLMs for Robotics
2024-09-27TaskGen Tutorial 6: Conversation Wrapper
2024-09-26TaskGen Tutorial 5: External Functions & CodeGen