claim
active
claim:imtl-l-is-equivalent-to-the-logarithm-transformation-when-its-parameter-st-is-the-exact-minimizer-in-each-iterationIMTL-L is equivalent to the logarithm transformation when its parameter st is the exact minimizer in each iteration.
Mathematical relationship between IMTL-L and log transformation.
Source paper
extracted_from(2023) · Baijiong Lin · Weisen Jiang · Feiyang Ye · Yu Zhang +5
Neighborhood — ranked by edge-count
Communities (3)
community
- Dual-balancing multi-task learningmembers_ofDB-MTL jointly balances loss scale and gradient magnitude, benchmarked on NYUv2 and Office-31.
- Dual balancing multi-task learningmembers_ofDB-MTL combines loss-scale and gradient-magnitude balancing, benchmarked across NYUv2, Cityscapes, QM9, and Office datasets.
- Parameter-free logarithm transformation for multi-task learning that improves gradient balancing methods like PCGrad and Nash-MTL across vision benchmarks.
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- The logarithm transformation is simpler and more effective than IMTL-L because it is parameter-free.claim0.882Comparison of loss-scale balancing techniques.
- Comparison of loss-scale balancing with IMTL-L.
- Prior loss-balancing method using learnable loss transformation; logarithm approach recovers this
- Gradient balancing enforcing equal projections on each task gradient.
- Effectiveness of logarithm transformation as a plug-in for gradient balancing methods.
- Concise summary of the DB-MTL method from the abstract.
- Compared to IMTL-L: parameter-free, no extra computational cost, achieves same theoretical goal
- We find that the logarithm transformation also benefits existing gradient balancing methods.quote0.751Key finding showing the broader utility of the log transformation.