Premium Content

Adaptive MTL Baselines

Fixed weighting, Homoscedastic Uncertainty, GradNorm, and DWA — the published baselines we ran inside the same framework.

This chapter requires a subscription to access.

What you'll unlock:

  • 1. Fixed Equal Weighting (0.5 / 0.5)
  • 2. Homoscedastic Uncertainty (Kendall et al.)
  • 3. GradNorm — and Why It Diverged on N-CMAPSS
  • 4. Dynamic Weight Average (DWA)
  • 5. Implementation & Results Summary
Subscribe to Unlock

Already have an account? Sign in