On the Benefits of Invariance in Neural Networks.¶
Lyle, C., Wilk, v. d., Kwiatkowska, Gal, & Bloem-Reddy, B. (2020). On the Benefits of Invariance in Neural Networks. arXiv preprint.
Cited by¶
1 citation across 1 artifact.
Each citation links to the sentence it supports in the citing article.
Primes¶
- Equivariance
- The prime names the relation; it does not adjudicate the modeling choice, a caution Lyle and colleagues (2020) raise when they show invariance and equivariance help only when the assumed symmetry genuinely holds in the data.
This sourceAnalyzes when building in invariance/equivariance helps (data augmentation and feature averaging), proving generalization benefits under the assumed invariant structure.
- The prime names the relation; it does not adjudicate the modeling choice, a caution Lyle and colleagues (2020) raise when they show invariance and equivariance help only when the assumed symmetry genuinely holds in the data.
Verification¶
This reference passed the adversarial substantiation pipeline: it was checked to exist and to support the claim it is attached to. See how references were verified.
Registry ID ref:d7684645b198 · see in the full table