On the Informativeness of Supervision Signals

Ilia Sucholutsky, Ruairidh M. Battleday, Katherine M. Collins, Raja Marjieh, Joshua C. Peterson, Pulkit Singh, Umang Bhatt, Nori Jacoby, Adrian Weller, Thomas L. Griffiths

Research output: Contribution to journalConference articlepeer-review


Supervised learning typically focuses on learning transferable representations from training examples annotated by humans. While rich annotations (like soft labels) carry more information than sparse annotations (like hard labels), they are also more expensive to collect. For example, while hard labels only provide information about the closest class an object belongs to (e.g., “this is a dog”), soft labels provide information about the object's relationship with multiple classes (e.g., “this is most likely a dog, but it could also be a wolf or a coyote”). We use information theory to compare how a number of commonly-used supervision signals contribute to representation-learning performance, as well as how their capacity is affected by factors such as the number of labels, classes, dimensions, and noise. Our framework provides theoretical justification for using hard labels in the big-data regime, but richer supervision signals for few-shot learning and out-of-distribution generalization. We validate these results empirically in a series of experiments with over 1 million crowdsourced image annotations and conduct a cost-benefit analysis to establish a tradeoff curve that enables users to optimize the cost of supervising representation learning on their own datasets.

Original languageEnglish (US)
Pages (from-to)2036-2046
Number of pages11
JournalProceedings of Machine Learning Research
StatePublished - 2023
Event39th Conference on Uncertainty in Artificial Intelligence, UAI 2023 - Pittsburgh, United States
Duration: Jul 31 2023Aug 4 2023

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence
  • Software
  • Control and Systems Engineering
  • Statistics and Probability


Dive into the research topics of 'On the Informativeness of Supervision Signals'. Together they form a unique fingerprint.

Cite this