all AI news
Constraining Representations Yields Models That Know What They Don't Know. (arXiv:2208.14488v2 [cs.LG] UPDATED)
Sept. 30, 2022, 1:16 a.m. | Joao Monteiro, Pau Rodriguez, Pierre-Andre Noel, Issam Laradji, David Vazquez
cs.CV updates on arXiv.org arxiv.org
A well-known failure mode of neural networks is that they may confidently
return erroneous predictions. Such unsafe behaviour is particularly frequent
when the use case slightly differs from the training context, and/or in the
presence of an adversary. This work presents a novel direction to address these
issues in a broad, general manner: imposing class-aware constraints on a
model's internal activation patterns. Specifically, we assign to each class a
unique, fixed, randomly-generated binary vector - hereafter called class code - …
More from arxiv.org / cs.CV updates on arXiv.org
Jobs in AI, ML, Big Data
Senior ML Researcher - 3D Geometry Processing | 3D Shape Generation | 3D Mesh Data
@ Promaton | Europe
Analytics Engineer
@ CircleCI | Remote (US), Remote (Canada), San Francisco, Denver
Bilingual Executive Assistant/Data Analyst - (French and English) - Export
@ Dangote Group | Lagos, Lagos, Nigeria
Workday Services Data Lead
@ WPP | Mexico City, Mexico
Business Data Analyst
@ Nordea | Tallinn, EE, 11415
Data Integrity Lead
@ BioNTech SE | Gaithersburg, MD, US, MD 20878