DB-MTL reduces gradient conflict and improves task balance compared to EW.

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• DB-MTL achieves loss-scale balancing by performing logarithm transformation on each task loss, and rescales gradient magnitudes by normalizing all task gradients to comparable magnitudes using the maximum gradient norm.quote0.848
  Concise summary of the DB-MTL method from the abstract.
• DB-MTL is a simple yet effective method that addresses both loss-scale and gradient-magnitude imbalances.claim0.839
  Core claim of the paper.
• DB-MTL has similar per-epoch running time to gradient balancing methods on NYUv2, slower than loss balancing methods.finding0.826
  Computational efficiency comparison.
• DB-MTL does not affect training stability; losses converge smoothly.claim0.821
  Training stability claim.
• DB-MTL training losses decrease smoothly and gradient norms are lower than EW on NYUv2, indicating training stability.finding0.809
  Training stability analysis.
• As can be seen, most MTL baselines perform better than STL on semantic segmentation and depth estimation, but have a large drop on the surface normal prediction task, suffering from the task balancing problem.quote0.792
  Observation illustrating the task balancing problem on NYUv2.
• DB-MTL with EMA forgetting rate β in a wide range performs better than without EMA (β=0) on Office-31.finding0.779
  Effect of EMA forgetting rate on performance.
• DB-MTL (Dual-Balancing Multi-Task Learning)method0.776
  The proposed method combining loss-scale balancing via logarithm transformation and gradient-magnitude balancing via maximum-norm normalization.