Learning Bounds for Domain Adaptation

By John Blitzer, Koby Crammer, Alex Kulesza, Fernando C Pereira, Jennifer Wortman Vaughan. 2007

532 citations recorded by the source corpus.

Subfields: Representation & Self-Supervised Learning, Statistical Learning Theory

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

In Statistical Learning Theory

Citations in the corpus

A theory of learning from different domains (2010, methodology): “Preliminary versions of the work contained in this article appeared in Advances in Neural Information Processing Systems (Ben-David et al. 2006; Blitzer et al. 2007a ).”
Multiple-Source Adaptation with Domain Classifiers (2020, background): “The general theoretical problem of adaptation from a single domain to a target domain has been studied in a series of publications in the last two decades or so [Kifer et al., 2004, Ben-David et al., 2007, Blitzer et al., 2008, Mansour et al., 2009b, Cortes and Mohri, 2014, Cortes et al., 2019b].”
A Theory of Multiple-Source Adaptation with Limited Target Labeled Data (2020, background): “The single-source adaptation problem was first studied by Ben-David et al. (2007), as well as follow-up publications (Blitzer et al., 2008; Ben-David et al., 2010), where the authors presented an analysis in terms of a dA-distance, including VC-dimension learning bounds for the zero-one loss.”
Three Approaches for Personalization with Applications to Federated Learning (2020, methodology): “As in domain adaptation with target labels (Blitzer et al., 2008), we have at our disposal a large amount of labeled data from the source (global data) and a small amount of labeled data from the target (personal data).”
Domain Adaptation: Learning Bounds and Algorithms (2009, methodology): “The problem of domain adaptation arises in a variety of applications in natural language processing (Dredze et al., 2007; Blitzer et al., 2007; Jiang & Zhai, 2007; Chelba & Acero, 2006; Daumé III & Marcu, 2006), speech processing (Legetter & Woodland, 1995; Gauvain & Chin-Hui, 1994; Pietra et al.,…”
Deep Domain Confusion: Maximizing for Domain Invariance (2014, background): “A number of recent theoretical and empirical results have shown that supervised methods’ test error increases in proportion to the difference between the test and training input distribution [3, 5, 29, 32].”
Moment Matching for Multi-Source Domain Adaptation (2018, neutral): “In contrast with existing theoretical analysis on domain adaptation [1, 57, 3] that is based on a divergence measure known as the H-divergence, we introduce the following definition of cross-moment divergence.”
An embarrassingly simple approach to zero-shot learning (2015, background): “Theorem 1 in (Blitzer et al., 2008) states that given a hypothesis space H of VC-dimension d̄, and sets U , U � of m̄ instances sampled i.”
Multiple Source Adaptation and the Rényi Divergence (2009, background): “Several recent studies deal with some theoretical aspects of this adaptation problem [2, 3].”
One-Shot Adaptation of Supervised Deep Convolutional Models (2013, background): “Theoretical [2, 4] and practical results from [20, 21] have shown that supervised methods’ test error increases in proportion to the difference between the test and training input distribution.”
Generalizing from a Few Examples (2019, background): “rt from that, as FSL sometimes becomes domain adaptation problems, such as those mentioned in multitask learning methods in Section 4.1. Existing bounds on estimation error E est(H,I)may be inspiring [10, 15]. Finally, convergence for methods in algorithms are not fully understood. Indeed, search us”
Scalable Transfer Learning with Expert Models (2020, background): “This can improve both convergence rates (Ben-David et al., 2007; 2010; Blitzer et al., 2008; Du et al., 2017; Kuzborskij & Orabona, 2013; Mansour et al., 2009) and empirical performance (Dai et al.”
Adaptation Algorithm and Theory Based on Generalized Discrepancy (2014, methodology): “LEARNING SCENARIO This section defines the learning scenario of domain adaptation we consider, which coincides with that of Blitzer et al. [2007a], Mansour et al. [2009a], or Cortes and Mohri [2013] and introduces the definitions and concepts needed for the following sections. For the most part, we ”
A Survey on Transfer Learning (2010, background): “Blitzer et al. [38] proposed to use SCL for solving NLP problems.”
Domain Adaptation with Coupled Subspaces (2011, background): “Either the source and target distributions share support [18, 20, 10], or they have low divergence for a specific hypothesis class [6, 28].”
Leveraging Sequence Classification by Taxonomy-Based Multitask Learning (2010, background): “A family of machine learning methods, commonly referred to as domain adaptation or transfer learning, investigates the application of a predictor trained with data from a given domain to data from a different one (see e.g., [ 3 ,5,11]).”
Federated Learning for Open Banking (2021, methodology): “Another solution is to treat each customer as a target task and the pre-trained global model as a source task, then to apply transfer learning [28] or domain adaptation [3] methods to ﬁne-tune each personalized model.”
Machine Learning Paradigms for Speech Recognition: An Overview (2013, unknown): “The work by [185]”