Terra Nova: A Comprehensive Challenge Environment for Intelligent Agents
Fuente:
arXiv
Gespeichert in:
| 1. Verfasser: | |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866911276529090560 |
|---|---|
| author | McInroe, Trevor |
| author_facet | McInroe, Trevor |
| contents | We introduce Terra Nova, a new comprehensive challenge environment (CCE) for reinforcement learning (RL) research inspired by Civilization V. A CCE is a single environment in which multiple canonical RL challenges (e.g., partial observability, credit assignment, representation learning, enormous action spaces, etc.) arise simultaneously. Mastery therefore demands integrated, long-horizon understanding across many interacting variables. We emphasize that this definition excludes challenges that only aggregate unrelated tasks in independent, parallel streams (e.g., learning to play all Atari games at once). These aggregated multitask benchmarks primarily asses whether an agent can catalog and switch among unrelated policies rather than test an agent's ability to perform deep reasoning across many interacting challenges. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_15378 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Terra Nova: A Comprehensive Challenge Environment for Intelligent Agents McInroe, Trevor Artificial Intelligence We introduce Terra Nova, a new comprehensive challenge environment (CCE) for reinforcement learning (RL) research inspired by Civilization V. A CCE is a single environment in which multiple canonical RL challenges (e.g., partial observability, credit assignment, representation learning, enormous action spaces, etc.) arise simultaneously. Mastery therefore demands integrated, long-horizon understanding across many interacting variables. We emphasize that this definition excludes challenges that only aggregate unrelated tasks in independent, parallel streams (e.g., learning to play all Atari games at once). These aggregated multitask benchmarks primarily asses whether an agent can catalog and switch among unrelated policies rather than test an agent's ability to perform deep reasoning across many interacting challenges. |
| title | Terra Nova: A Comprehensive Challenge Environment for Intelligent Agents |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2511.15378 |