Matches in SemOpenAlex for { <https://semopenalex.org/work/W4226412614> ?p ?o ?g. }
Showing items 1 to 65 of
65
with 100 items per page.
- W4226412614 abstract "Currently, it is hard to reap the benefits of deep learning for Bayesian methods, which allow the explicit specification of prior knowledge and accurately capture model uncertainty. We present Prior-Data Fitted Networks (PFNs). PFNs leverage large-scale machine learning techniques to approximate a large set of posteriors. The only requirement for PFNs to work is the ability to sample from a prior distribution over supervised learning tasks (or functions). Our method restates the objective of posterior approximation as a supervised classification problem with a set-valued input: it repeatedly draws a task (or function) from the prior, draws a set of data points and their labels from it, masks one of the labels and learns to make probabilistic predictions for it based on the set-valued input of the rest of the data points. Presented with a set of samples from a new supervised learning task as input, PFNs make probabilistic predictions for arbitrary other data points in a single forward propagation, having learned to approximate Bayesian inference. We demonstrate that PFNs can near-perfectly mimic Gaussian processes and also enable efficient Bayesian inference for intractable problems, with over 200-fold speedups in multiple setups compared to current methods. We obtain strong results in very diverse areas such as Gaussian process regression, Bayesian neural networks, classification for small tabular data sets, and few-shot image classification, demonstrating the generality of PFNs. Code and trained PFNs are released at https://github.com/automl/TransformersCanDoBayesianInference." @default.
- W4226412614 created "2022-05-05" @default.
- W4226412614 creator A5018199868 @default.
- W4226412614 creator A5031002895 @default.
- W4226412614 creator A5052679852 @default.
- W4226412614 creator A5055300090 @default.
- W4226412614 creator A5081550394 @default.
- W4226412614 date "2021-12-20" @default.
- W4226412614 modified "2023-09-24" @default.
- W4226412614 title "Transformers Can Do Bayesian Inference" @default.
- W4226412614 doi "https://doi.org/10.48550/arxiv.2112.10510" @default.
- W4226412614 hasPublicationYear "2021" @default.
- W4226412614 type Work @default.
- W4226412614 citedByCount "0" @default.
- W4226412614 crossrefType "posted-content" @default.
- W4226412614 hasAuthorship W4226412614A5018199868 @default.
- W4226412614 hasAuthorship W4226412614A5031002895 @default.
- W4226412614 hasAuthorship W4226412614A5052679852 @default.
- W4226412614 hasAuthorship W4226412614A5055300090 @default.
- W4226412614 hasAuthorship W4226412614A5081550394 @default.
- W4226412614 hasBestOaLocation W42264126141 @default.
- W4226412614 hasConcept C107673813 @default.
- W4226412614 hasConcept C119857082 @default.
- W4226412614 hasConcept C121332964 @default.
- W4226412614 hasConcept C136389625 @default.
- W4226412614 hasConcept C153083717 @default.
- W4226412614 hasConcept C154945302 @default.
- W4226412614 hasConcept C160234255 @default.
- W4226412614 hasConcept C163716315 @default.
- W4226412614 hasConcept C2776214188 @default.
- W4226412614 hasConcept C41008148 @default.
- W4226412614 hasConcept C49937458 @default.
- W4226412614 hasConcept C50644808 @default.
- W4226412614 hasConcept C61326573 @default.
- W4226412614 hasConcept C62520636 @default.
- W4226412614 hasConceptScore W4226412614C107673813 @default.
- W4226412614 hasConceptScore W4226412614C119857082 @default.
- W4226412614 hasConceptScore W4226412614C121332964 @default.
- W4226412614 hasConceptScore W4226412614C136389625 @default.
- W4226412614 hasConceptScore W4226412614C153083717 @default.
- W4226412614 hasConceptScore W4226412614C154945302 @default.
- W4226412614 hasConceptScore W4226412614C160234255 @default.
- W4226412614 hasConceptScore W4226412614C163716315 @default.
- W4226412614 hasConceptScore W4226412614C2776214188 @default.
- W4226412614 hasConceptScore W4226412614C41008148 @default.
- W4226412614 hasConceptScore W4226412614C49937458 @default.
- W4226412614 hasConceptScore W4226412614C50644808 @default.
- W4226412614 hasConceptScore W4226412614C61326573 @default.
- W4226412614 hasConceptScore W4226412614C62520636 @default.
- W4226412614 hasLocation W42264126141 @default.
- W4226412614 hasOpenAccess W4226412614 @default.
- W4226412614 hasPrimaryLocation W42264126141 @default.
- W4226412614 hasRelatedWork W2110658950 @default.
- W4226412614 hasRelatedWork W2338752163 @default.
- W4226412614 hasRelatedWork W2511279186 @default.
- W4226412614 hasRelatedWork W2746267661 @default.
- W4226412614 hasRelatedWork W2803667729 @default.
- W4226412614 hasRelatedWork W2806909761 @default.
- W4226412614 hasRelatedWork W2963058055 @default.
- W4226412614 hasRelatedWork W3040426824 @default.
- W4226412614 hasRelatedWork W96862169 @default.
- W4226412614 hasRelatedWork W4286903091 @default.
- W4226412614 isParatext "false" @default.
- W4226412614 isRetracted "false" @default.
- W4226412614 workType "article" @default.