- Title
- Target curricula via selection of minimum feature sets: a case study in Boolean networks
- Creator
- Fenn, Shannon; Moscato, Pablo
- Relation
- Funding BodyARCGrant NumberFT120100060 and ARC DP140104183 http://purl.org/au-research/grants/arc/DP140104183
- Relation
- Journal of Machine Learning Research Vol. 18, p. 1-26
- Relation
- http://jmlr.org/papers/v18/17-007.html
- Publisher
- MIT Press
- Resource Type
- journal article
- Date
- 2017
- Description
- We consider the effect of introducing a curriculum of targets when training Boolean models on supervised Multi Label Classification (MLC) problems. In particular, we consider how to order targets in the absence of prior knowledge, and how such a curriculum may be enforced when using meta-heuristics to train discrete non-linear models. We show that hierarchical dependencies between targets can be exploited by enforcing an appropriate curriculum using hierarchical loss functions. On several multi-output circuit-inference problems with known target difficulties, Feedforward Boolean Networks (FBNs) trained with such a loss function achieve significantly lower out-of-sample error, up to 10% in some cases. This improvement increases as the loss places more emphasis on target order and is strongly correlated with an easy-to-hard curricula. We also demonstrate the same improvements on three real-world models and two Gene Regulatory Network (GRN) inference problems. We posit a simple a-priori method for identifying an appropriate target order and estimating the strength of target relationships in Boolean MLCs. These methods use intrinsic dimension as a proxy for target dificulty, which is estimated using optimal solutions to a combinatorial optimisation problem known as the Minimum-Feature-Set (minFS) problem. We also demonstrate that the same generalisation gains can be achieved without providing any knowledge of target difficulty.
- Subject
- multi-label classification; target curriculum; Boolean networks; k-feature set
- Identifier
- http://hdl.handle.net/1959.13/1357624
- Identifier
- uon:31944
- Identifier
- ISSN:1532-4435
- Rights
- Copyright 2017 Shannon Fenn and Pablo Moscato. License: CC-BY 4.0, see https://creativecommons.org/licenses/by/4.0/. Attribution requirements are provided at http://jmlr.org/papers/v18/17-007.html.
- Language
- eng
- Full Text
- Reviewed
- Hits: 1476
- Visitors: 1820
- Downloads: 356
Thumbnail | File | Description | Size | Format | |||
---|---|---|---|---|---|---|---|
View Details Download | ATTACHMENT02 | Publisher version (open access) | 578 KB | Adobe Acrobat PDF | View Details Download |