finding
active
finding:log-x-min-s-e-s-x-s-1-for-x-0log(x) = min_s (e^s * x - s - 1) for x > 0
Mathematical equivalence showing logarithm transformation recovers IMTL-L in the limit
Source paper
extracted_from(2023) · Baijiong Lin · Weisen Jiang · Feiyang Ye · Yu Zhang +5
Neighborhood — ranked by edge-count
Claims (1)
claim
- Compared to IMTL-L: parameter-free, no extra computational cost, achieves same theoretical goal
Communities (2)
community
- Dual-balancing multi-task learningmembers_ofDB-MTL jointly balances loss scale and gradient magnitude, benchmarked on NYUv2 and Office-31.
- 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.
- Mathematical relationship between IMTL-L and log transformation.
- Justification for the linear combination
- A heuristic measure of degree of life in buildings, combining harmony (H) and temperature (T) to approximate the density of living centers.
- The logarithm transformation is simpler and more effective than IMTL-L because it is parameter-free.claim0.728Comparison of loss-scale balancing techniques.
- Steering vectors from µ(0→2) slightly outperform µ(1→2) for instruction discovery across datasets and modelsfinding0.725Shows that contrasting No Reflection with Triggered Reflection provides a stronger signal than Intrinsic vs Triggered.
- E3 finding distinguishing the two geometry summaries; breadth less predictive than peak height
- S = ρd − dr − log k is a predictive correlate of anchoring success across few-shot, SFT, and CoT.claim0.703UCCT's practical utility claim.
- Parameter-free loss transformation applied to each task loss to equalize scales