wenmin-wu/timeseries-mc-dropout-uncertainty

MC Dropout Uncertainty

Overview

Monte Carlo Dropout keeps dropout active at inference time and runs N forward passes. The mean gives a smoothed prediction; the standard deviation gives a calibrated uncertainty estimate. No architectural changes needed — just force training=True on dropout layers.

Quick Start

import numpy as np

def mc_dropout_predict(model, X, n_samples=100):
    """Run N stochastic forward passes and return mean + std."""
    preds = np.stack([model(X, training=True).numpy() for _ in range(n_samples)])
    return preds.mean(axis=0), preds.std(axis=0)

# PyTorch version
def mc_predict_torch(model, X, n_samples=100):
    model.train()  # keeps dropout active
    with torch.no_grad():
        preds = torch.stack([model(X) for _ in range(n_samples)])
    return preds.mean(0), preds.std(0)

Key Decisions

• n_samples=100: good balance of quality vs compute; 25+ gives usable uncertainty
• training=True: the only change needed — forces dropout to sample
• Quadrature propagation: combine MC std with other uncertainty sources via sqrt(σ1² + σ2²)
• Works with any dropout model: no special architecture required

References

• Source: host-starter-solution
• Competition: NeurIPS - Ariel Data Challenge 2024