Any learning algorithm must carry bias toward particular targets, with bias acting as a conserved quantity that creates strict trade-offs across tasks.
In: Proc eedings of the 13th International Conference on Neural Information Processin g Systems
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
The Futility of Bias-Free Learning and Search
Any learning algorithm must carry bias toward particular targets, with bias acting as a conserved quantity that creates strict trade-offs across tasks.