Local Learning Algorithms

By L. Eon Bottou, Vladimir Naumovich Vapnik. 1992

How researchers see this paper

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

Citations in the corpus

Learning Explainable Decision Rules via Maximum Satisfiability (2020, neutral): “learning approach for query-specific applications [17], [18].”
Recasting Gradient-Based Meta-Learning as Hierarchical Bayes (2018, background): “…in a hierarchical Bayesian model that regularize task-speciﬁc parameters (Heskes, 1998), as a learned metric space in which to group neighbors (Bottou & Vapnik, 1992), as a trained recurrent neural network that allows encoding and retrieval of episodic information (Santoro et al., 2016), or…”
Transformation Invariance in Pattern Recognition-Tangent Distance and Tangent Propagation (1996, methodology): “Other algorithms such as Neural nets [18, 20], Optimal Margin Classi er [7], Local Learning [3] and Boosting [9] were also used on these databases.”
Incremental Support Vector Machine Learning: A Local Approach (2001, methodology): “In some extent, the proposed algorithm is linked with work of Bottou and Vapnik [1] about local learning algorithm.”
Fine-Grained Visual Comparisons with Local Learning (2014, methodology): “Furthermore, while local learning methods have been explored for classification [2, 6, 17, 31, 33] and information retrieval problems [3, 13, 16, 24], our idea for learning local ranking functions with comparisons is new.”
Transformation invariance in pattern recognition: Tangent distance and propagation (2000, methodology): “Other algorithms such as neural nets (LeCun et al., 1990, 1995), optimal margin classifier (Cortes and Vapnik, 1995), local learning (Bottou and Vapnik, 1992), and boosting (Drucker et al., 1993) were also used on these databases.”
Supervised Learning in Spiking Neural Networks with ReSuMe: Sequence Learning, Classification, and Spike Shifting (2010, background): “In this case, the system should reject a pattern whenever the classiﬁcation cannot be achieved with enough conﬁdence (see,e.g.,Bottou&Vapnik,1992,fordiscussion).”
Machine Learning With Big Data: Challenges and Approaches (2017, background): “First proposed by Bottou and Vapnik in 1992 [109], local learning is a strategy that offers an alternative to typical global learning.”
Efficient Machine Learning for Big Data: A Review (2015, background): “A number of recent works have demonstrated that such a local learning strategy is far superior to that of the global learning strategy, especially on data sets that are not evenly distributed [37–40].”
Classification in the Presence of Label Noise: A Survey (2014, background): “Local models [174] can also be used to ﬁlter mislabeled training samples.”
Learning to (Learn at Test Time): RNNs with Expressive Hidden States (2024, unknown): “One of the earliest versions of this idea is called local learning (Bottou and Vapnik [9]): For each test input, train on its neighbors before making a prediction.”
Vicinal Risk Minimization (2000, background): “…Set Hard SVM VRM Fixed slope VRM Adaptive slope 250+250 3.34% 2.79% 2.54% 500+500 3.11% 2.47% 2.27% 1000+1000 2.94% 2.08% 1.96% The statistical signifiance of these results can be asserted with very high probability by comparing the list of errors performed by each system (Bottou and Vapnik, 1992).”