❶ A neural network has been trained to create spatial maps and predict subsequent video frames using a predictive coding algorithm and Minecraft gameplay, achieving a mean-squared error of 0.094%. ❷ The project, led by researchers at Caltech, demonstrates AI's ability to navigate conceptual spaces, moving beyond simple input-response memorization. ❸ The research, published in Nature Machine Intelligence, includes detailed methodology and the code is available on GitHub and Zenodo for public use.
Related Articles
- Minecraft is finally getting a massive visual overhaul in upcoming update4 months ago
- Novel Memristors to Combat the 'Catastrophic Forgetting' of AI4 months ago
- HM25: Smart Systems in Their Future Applications - Reliable, Trustworthy, and Secure4 months ago
- JFrog: A Tech Stock Leaps And Bounds Above Others4 months ago
- University Satellite SONATE-2 in Orbit For a Year5 months ago
- Uni-Satellit SONATE-2 Completes One Year in Space5 months ago
- An Extraordinary Employee: Humanoid Robot TALOS at TU Darmstadt for Research into Human-like Learning5 months ago
- A Minecraft Movie gets ‘Final Trailer’ ahead of April premiere5 months ago
- Fraunhofer HHI Launches Research Project for Sustainable Communication Networks5 months ago
- ISSCC: CPUs and image processors5 months ago