TextWorldExpress: Simulating Text Games at One Million Steps Per Second

@article{Jansen2022TextWorldExpressST,
  title={TextWorldExpress: Simulating Text Games at One Million Steps Per Second},
  author={Peter Alexander Jansen and Marc-Alexandre C{\^o}t{\'e}},
  journal={ArXiv},
  year={2022},
  volume={abs/2208.01174}
}
Text-based games offer a challenging test bed to evaluate virtual agents at language understanding, multi-step problem-solving, and common-sense reasoning. However, speed is a major limitation of current text-based games, capping at 300 steps per second, mainly due to the use of legacy tooling. In this work we present TextWorldExpress, a high-performance simulator that includes implementations of three common text game benchmarks that increases simulation throughput by approximately three… 
[PDF] Semantic Reader
1 Citations
Methods Citations
1

Figures and Tables from this paper

One Citation

Behavior Cloned Transformers are Neurosymbolic Reasoners

This experimental study indicates that injecting the actions from these symbolic modules into the action space of a behavior cloned transformer agent increases performance on four text game benchmarks that test arithmetic, navigation, sorting, and common sense reasoning by an average of 22%, allowing an agent to reach the highest possible performance on unseen games.

References

SHOWING 1-10 OF 40 REFERENCES

ScienceWorld: Is your Agent Smarter than a 5th Grader?

We present ScienceWorld, a benchmark to test agents’ scientific reasoning abilities in a new interactive text environment at the level of a standard elementary school science curriculum. Despite the…

Interactive Fiction Games: A Colossal Adventure

This work argues that IF games are an excellent testbed for studying language-based autonomous agents and introduces Jericho, a learning environment for man-made IF games and conducts a comprehensive study of text-agents across a rich set of games, highlighting directions in which agents can improve.

Efficient Text-based Reinforcement Learning by Jointly Leveraging State and Commonsense Graph Representations

This paper proposes an agent for TBGs that induces a graph representation of the game state and jointly grounds it with a graph of commonsense knowledge from ConceptNet, and shows that agents that incorporate commonsense into the gameState graph outperform baseline agents.

Megaverse: Simulating Embodied Agents at One Million Experiences per Second

The efficient design of the engine enables physics-based simulation with highdimensional egocentric observations at more than 1,000,000 actions per second on a single 8-GPU node, thereby taking full advantage of the massive parallelism of modern GPUs.

Reading and Acting while Blindfolded: The Need for Semantics in Text Game Agents

Surprisingly, an agent is capable of achieving high scores even in the complete absence of language semantics, indicating that the currently popular experimental setup and models may be poorly designed to understand and leverage game texts.

Text-based RL Agents with Commonsense Knowledge: New Challenges, Environments and Baselines

This paper designs a new text-based gaming environment called TextWorld Commonsense (TWC) for training and evaluating RL agents with a specific kind of commonsense knowledge about objects, their attributes, and affordances, and shows that agents which incorporate Commonsense knowledge in TWC perform better, while acting more efficiently.

How to Motivate Your Dragon: Teaching Goal-Driven Agents to Speak and Act in Fantasy Worlds

A reinforcement learning system is introduced that incorporates large-scale language modeling-based and commonsense reasoning-based pre-training to imbue the agent with relevant priors and leverages a factorized action space of action commands and dialogue, balancing between the two.

The NetHack Learning Environment

It is argued that NetHack is sufficiently complex to drive long-term research on problems such as exploration, planning, skill acquisition, and language-conditioned RL, while dramatically reducing the computational resources required to gather a large amount of experience.

ALFRED: A Benchmark for Interpreting Grounded Instructions for Everyday Tasks

It is shown that a baseline model based on recent embodied vision-and-language tasks performs poorly on ALFRED, suggesting that there is significant room for developing innovative grounded visual language understanding models with this benchmark.

Special Feature Zork: A Computerized Fantasy Simulation Game

In the demonic world of Zork, a simulated universe entices the player into a new form of problem solving.