# Visual Rl Rectification Eval

> Evaluates the stability and generalization of PPO policies trained with mode-dependent layers (BatchNorm, dropout) across visual reinforcement learning environments. It probes whether a deterministic rectification phase prevents reward collapse and aligns training-evaluation dynamics compared to standard training modes. Use when the user wants to benchmark on Procgen, Histopathology Patch-Localization, Natural Image Patch-Localization, or asks about evaluating this task. Reports normalized reward (%).

- Skill: `qhjqhj00/visual-rl-rectification-eval` (Agent Skill)
- Install (CLI): `npx skillmds add qhjqhj00/visual-rl-rectification-eval`
- Raw SKILL.md: https://api.skillmd.com/api/skills/qhjqhj00/visual-rl-rectification-eval/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Productivity
- Author: qhjqhj00 (https://skillmd.com/u/qhjqhj00)
- Updated: 2026-09-08
- Page: https://skillmd.com/skills/qhjqhj00/visual-rl-rectification-eval

---


# visual-rl-rectification-eval

> Mode-Dependent Rectification for Stable PPO Training — Mohamad, Ponzio, and Descombes (2026) (arXiv:2602.05619, 2026)

## What this evaluates

Evaluates the stability and generalization of PPO policies trained with mode-dependent layers (BatchNorm, dropout) across visual reinforcement learning environments. It probes whether a deterministic rectification phase prevents reward collapse and aligns training-evaluation dynamics compared to standard training modes.

## Datasets

- **Procgen** — total 1000; splits: train (500), test (500)
- **Histopathology Patch-Localization** — total ?; splits: train (256), train (1024)
- **Natural Image Patch-Localization** — total ?; splits: train (256), train (1024)

## Metrics

- `normalized reward (%)` **(primary)** — range: percent
  - Episode return divided by the optimal policy reward, multiplied by 100 to express performance as a percentage of the theoretical maximum.
- `average episode return` — range: other
  - Mean sum of rewards collected per episode across test environments, reported as Score.

## Input / output format

**Input**: RGB image observations from the environment, processed through a shallow ResNet-18 backbone initialized with ImageNet weights.

**Output**: Action distribution sampled via the PPO policy network (discrete or continuous depending on the environment).

## Scoring recipe

```python
def compute_metrics(returns, optimal_rewards):
    normalized_reward_pct = (returns / optimal_rewards) * 100
    avg_episode_return = np.mean(returns)
    return normalized_reward_pct, avg_episode_return
```

## Common pitfalls

- Standard BatchNorm training causes severe reward collapse on high-variability visual tasks, making direct baseline comparisons impossible without rectification.
- Evaluation must strictly separate training-mode and evaluation-mode statistics for mode-dependent layers, as their non-stationary distributions diverge during PPO optimization.
- Procgen results are averaged over 500 easy levels for training and 500 unseen levels for testing; mixing these splits invalidates generalization claims.

## Evidence (verbatim from paper)

> Figure 3 (top) reports performance on the patch-localization tasks, expressed as a percentage of the optimal policy reward. Figure 3 (bottom) reports performance on the Progen benchmark, measured as average episode return.

## Citation

```bibtex
@misc{mohamad2026modedependentrectification,
  title={Mode-Dependent Rectification for Stable PPO Training},
  author={Mohamad, Ponzio, and Descombes (2026)},
  year={2026},
  note={arXiv:2602.05619}
}
```

- arXiv: 2602.05619