͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏

Forwarded this email? Subscribe here for more

Was this email forwarded to you? Sign up here

Edge 434: How Google DeepMind’s GameNGen can Simulate Entire 1993’s DOOM Game in Real Time

A major milestone in creating generative AI models that can interact with complex real world environments.

Sep 26

READ IN APP

Simulating and interacting with real-world environments is one of the next frontiers for generative AI models. From robotics to self-driving vehicles, there are countless of scenarios that require generative AI agents to perform actions in real-world environments. Part of the challenges when building these models has to do with the size of the training corpus. Collecting data from real-world environments is particularly complex and expensive. But can we create generative AI agents that simulate entire environments with high fidelity? This is the goal of a new research from Google DeepMind that showcased GameNGen, pronounced as “game engine”, a model that can simulate a DOOM game using a diffusion model.

GameNGen operates as a generative diffusion model designed to simulate video games. Traditional video games rely on a loop that gathers player inputs, updates the game state, and renders visuals on the screen. This loop runs at high speeds, creating the immersive experience that players enjoy. While these games usually run on standard computers, some unconventional hardware has been used to emulate games like DOOM, though the underlying software remains manually crafted. Regardless of the hardware, the core of game simulation involves predefined rules set by programmers.

Before we dive into GameNGen, check out the quality of simulations we are talking about:...

Subscribe to TheSequence to unlock the rest.

Become a paying subscriber of TheSequence to get access to this post and other subscriber-only content.

A subscription gets you:

	Full access to TheSequence Edge – what's new in AI + the most relevant ML concepts, research papers, tech solutions
	Full archive
	Comments and discussions

Like

Comment

Restack

Edge 434: How Google DeepMind’s GameNGen can Simulate Entire 1993’s DOOM Game in Real Time

Edge 434: How Google DeepMind’s GameNGen can Simulate Entire 1993’s DOOM Game in Real Time

A major milestone in creating generative AI models that can interact with complex real world environments.

Subscribe to TheSequence to unlock the rest.

A subscription gets you:

Older messages

Edge 433: Samba, Unlimited Context Windows and State Space Models

The Big Bucks in Gen AI Investments

Edge 432: NVIDIA Created Minitron by Distilling Llama 3.1

Edge 431: Meet the Multimodal State Space Models

Some Non-Obvious Points About OpenAI 01

You Might Also Like

Recording: 'Data Storytelling: What Organizations Need to Know Going Into 2025'

💻 Issue 437 - Introducing local Azure Service Bus Emulator

💎 Issue 444 - Why did people rub snow on frozen feet? (2017)

💻 Issue 444 - JavaScript Dos and Donts

📱 Issue 438 - Reverse Engineering iOS 18 Inactivity Reboot

💻 Issue 362 - React Anti-Pattern: Stop Passing Setters Down the Components Tree

💻 Issue 444 - Building simple event-driven applications with Pub/Sub

📱 Issue 441 - Shift Left Is the Tip of the Iceberg

💻 Issue 439 - Async/Await Is Real And Can Hurt You

📲 Why I Ditched Linux for Samsung DeX — Buy This Instead of a Gaming Headset