AI - AI Dynamics

Foundation Models Enable Automatic Curriculum Learning in RL

By

–

03 June 2023 19h31

RL is too difficult to learn from scratch. The future is large foundation models with world knowledge helping with automatic curriculum generation and online learning

MineCLIP Embeddings Open-Sourced on MineDojo

By

–

03 June 2023 19h28

Our MineCLIP embeddings are open-sourced on the Minedojo website. Fire away! https://
github.com/MineDojo

MineDojo: From Virtual Minecraft Worlds to Real-World AI Embodiment

By

–

03 June 2023 19h27

From #minecraft to the real world. That has been our goal when we open-sourced MineDojo. Embodiment in virtual world is the first-step. https://t.co/iF8DYqLwNP
— Prof. Anima Anandkumar (@AnimaAnandkumar) 3 juin 2023

From #minecraft to the real world. That has been our goal when we open-sourced MineDojo. Embodiment in virtual world is the first-step.

Minecraft Voyager: Embodied AI and Interactive Curriculum Learning

By

–

03 June 2023 19h25

Thank you @erikbryn Minecraft Voyager is an excited step towards building embodied #AI and getting #GPT4 to interactively build a curriculum and skill library

→ View original post on X — @yitayml

LLM Evaluation and Benchmarking: Critical Missing Piece

By

@yitayml

–

03 June 2023 19h15

Eval and good benchmarking for LLMs is one very big piece missing right now.

Minedojo Voyager Contributes New Skills to Minecraft Library

By

–

03 June 2023 19h12

Do contribute newly discovered skills into Minedojo Voyager skill library #minecraft #GPT4

→ View original post on X — @h2oai

Regulatory Framework for Ethical Risks in Generative AI

By

@h2oai

–

03 June 2023 19h04

What should be the regulatory process to minimize #ethical risks around #GenerativeAI? 1. Self-governance
2. Testing
3. Responsible action
4. Continuous risk assessment What would you add? #chatGPT #responsibleAI

→ View original post on X — @nathanbenaich

Guide to AI Newsletter: Policy, LLM Research, and Startup Activity

By

@nathanbenaich

–

03 June 2023 18h41

Tomorrow’s issue of @guidetoai newsletter w/
@osebbouh is particularly rich in AI policy commentary…will AI exterminate the planet or not…and the latest LLM research and real-world product implementations. In addition to the usual summary of startup activity too 🙂 Sign