OSACT4 Reinforcement Learning Dataset

OSACT4 is a dataset for reinforcement learning tasks. The dataset originates from Kaggle, but its specific creator, size, and temporal details are not documented.

Use Cases

• Training agents to learn optimal policies based on state-space and reward signals.
• Benchmarking algorithm performance on defined action-space and environment interactions.
• Analyzing agent behavior trajectories across episodes.

Strengths

• Dataset is hosted on Kaggle, a platform with established data sharing practices.
• Focuses on reinforcement learning, a specific and active research area.

Limitations

• No information on row count, column count, or data completeness.
• Unknown license restricts clarity on permissible use.
• No temporal coverage or update history is available.

Provenance

Source

Kaggle