IMTL-G

Gradient balancing enforcing equal projections on each task gradient.

Neighborhood — ranked by edge-count

artifact

Lin 2023 Dual-Balancing for Multi-Task Learning
mentions
The paper proposing the Dual-Balancing Multi-Task Learning method.

cosine ≥ 0.65 · no typed edge

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

IMTL-Lframework0.881
Prior loss-balancing method using learnable loss transformation; logarithm approach recovers this
IMTL-L is equivalent to the logarithm transformation when its parameter st is the exact minimizer in each iteration.claim0.771
Mathematical relationship between IMTL-L and log transformation.
The logarithm transformation is simpler and more effective than IMTL-L because it is parameter-free.claim0.724
Comparison of loss-scale balancing techniques.
Googleinstitute0.716
Murray Shanahan's part-time employer and provider of LLM technology.
Google Xinstitute0.714
G Techinstitute0.707
Parlogframework0.704
A concurrent logic language featured prominently in recent ACM literature; used as a primary comparison point for Linda.
Logarithm transformation improves PCGrad, GradVac, IMTL-G, CAGrad, Nash-MTL, and Aligned-MTL on NYUv2 (Figure 1).finding0.695
Effectiveness of logarithm transformation as a plug-in for gradient balancing methods.