Learning with a probabilistic teacher

By Ashok K. Agrawala. 1970

Subfields: Representation & Self-Supervised Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Representation & Self-Supervised Learning

Citations in the corpus

Snorkel: Rapid Training Data Creation with Weak Supervision (2017, background): “Another type of supervision is self-training [3,44] and co-training [6], which involves training a model or pair of models on data that they labeled themselves.”
From Softmax to Sparsemax: A Sparse Model of Attention and Multi-Label Classification (2016, background): “asses of problems: 3Not to be confused with “multi-class classiﬁcation,” which denotes problems where Y= [K] and K > 2. 4This scenario is also relevant for “learning with a probabilistic teacher” (Agrawala, 1970) and semi-supervised learning (Chapelle et al., 2006), as it can model label uncertai”
Active and Semisupervised Learning for the Classification of Remote Sensing Images (2014, background): “18 and 19 show the distribution of unlabeled and labeled samples at different iterations of the AL process for experiments 2 and 4, respectively.”
Semi-Supervised Learning (2019, unknown): “, Scudder [1965], Fralick [1967], Agrawala [1970]). An unsatisfactory aspect of self-learning is that the effect of the wrapper depends on the supervised method used inside it. If self-learning is used with empirical risk minimization and 1-0-loss, the unlabeled data will have no effect on the solut”
Snorkel DryBell: A Case Study in Deploying Weak Supervision at Industrial Scale (2018, background): “Methods in which a learner labels additional data for itself to train on include self-training [2, 34], co-training [5], and pseudo-labeling [18].”
Semi-Supervised Semantic Segmentation with Cross Pseudo Supervision (2021, background): “, self-learning, self-labeling, or decision-directed learning, is initially developed for using unlabeled data in classification [15, 10, 1, 3, 22].”
Snorkel: rapid training data creation with weak supervision (2019, background): “Another type of supervision is self-training [3,44] and co-training [6], which involves training a model or pair of models on data that they labeled themselves.”
Sensor fusion potential exploitation-innovative architectures and illustrative applications (1997, background): “It is to be noted in this context that earlier teacher models such as probabilistic teacher [ 11 ], which viewed the teacher as essentially neutral with an imperfectness of (where is the number of decision choices available) and imperfect teacher [12], which viewed the teacher as having a known leve”
Curriculum Labeling: Revisiting Pseudo-Labeling for Semi-Supervised Learning (2020, methodology): “Pseudo-labeling generally works by iteratively propagating labels from labeled samples to unlabeled samples using the current model to re-label the data (Scudder 1965; Fralick 1967; Agrawala 1970; Chapelle, Schlkopf, and Zien 2010).”
A Survey on Deep Semi-Supervised Learning (2021, background): “Since the 1970 s when the concept of SSL first came to the fore [6], [7], [8], there have been a wide variety of SSL meth-”