DSLE: A Learning Environment for Dark Souls Boss Encounters

Forum topic · 小凯 · 2026-08-11

Summary

The Dark Souls Learning Environment (DSLE) is a containerized platform exposing all 22 boss encounters of Dark Souls: Remastered as benchmarks for game-playing agents via a Gymnasium-style interface. Each environment step executes a real action against the running game, combining real-time combat, high-dimensional visual input, and sparse terminal rewards. The authors define DSLE-5, a representative five-boss subset covering melee combat, spatially constrained arenas, environmental hazards, multi-target fights, and fast final-boss encounters, recommended as a starting suite. On DSLE-5, they evaluate a random policy, an expert system, an evolutionary baseline, and PPO and DQN agents trained from visual input. The expert system and evolutionary baseline each defeated the tutorial boss (Asylum Demon) with peak win rates of 63% and 43% respectively, but none of the five methods defeated the other four bosses. PPO and DQN showed no measurable learning within budgets requiring tens of hours of wall-clock time per run (at most 0.33% win rate on the tutorial boss, 0% elsewhere). A broader evolutionary run across all 22 encounters, with fully maxed level-50 stats, won only a few additional early-game bosses. Failure modes are reported via survival time and damage dealt, ranging from deaths under 10 seconds in narrow multi-target fights to near-minute stalemates (arXiv:2508.03798).

Paper Overview

Research Area: Machine Learning Authors: Derin Gezgin, Jim O'Connor, Tanner Goodwin arXiv: 2508.03798

Abstract

We introduce the Dark Souls Learning Environment (DSLE), a containerized platform that presents all 22 boss encounters of Dark Souls: Remastered as game-playing agent benchmarks through a Gymnasium-style interface. DSLE combines real-time combat, high-dimensional visual input, and sparse terminal rewards, with each environment step being a real action executed against the running game. To support controlled comparison, the authors define DSLE-5, a representative five-boss subset spanning a melee fight, a spatially constrained arena, an environmental-hazard fight, a multi-target fight, and a fast final-boss fight, recommended as the starting suite for agents built on DSLE.

Key Results on DSLE-5

The authors evaluate five approaches: a random policy, an expert system, an evolutionary baseline, and PPO and DQN agents trained from visual input.

The expert system and evolutionary baseline each defeated the game's tutorial boss, the Asylum Demon, with peak win rates of 63% and 43% respectively.
None of the five methods defeated the other four bosses in DSLE-5.
PPO and DQN showed no measurable learning within evaluation budgets that already required tens of hours of wall-clock time per run: at most 0.33% win rate on the tutorial boss and 0% elsewhere.
A broader study ran the evolutionary baseline across all 22 encounters; even with fully maxed level-50 stats, it won only a handful of additional early-game bosses, losing the rest.

Failure Case Reporting

Failure modes are characterized using survival time and damage dealt rather than win rate alone. Observed failures range from deaths in under 10 seconds in narrow, multi-target encounters to stalemates lasting nearly a minute with almost no damage dealt.