Minimizing Statistical Bias with Queries

Abstract

I describe a querying criterion that attempts to minimize the error of a learner by minimizing its estimated squared bias. I describe experiments with locally-weighted regression on two simple prob(cid:173) lems, and observe that this "bias-only" approach outperforms the more common "variance-only" exploration approach, even in the presence of noise.

Cite

Text

Cohn. "Minimizing Statistical Bias with Queries." Neural Information Processing Systems, 1996.

Markdown

[Cohn. "Minimizing Statistical Bias with Queries." Neural Information Processing Systems, 1996.](https://mlanthology.org/neurips/1996/cohn1996neurips-minimizing/)

BibTeX

@inproceedings{cohn1996neurips-minimizing,
  title     = {{Minimizing Statistical Bias with Queries}},
  author    = {Cohn, David A.},
  booktitle = {Neural Information Processing Systems},
  year      = {1996},
  pages     = {417-423},
  url       = {https://mlanthology.org/neurips/1996/cohn1996neurips-minimizing/}
}