Multi-source domain adaptation and its application to early detection of fatigue

By Rita Chattopadhyay, Jieping Ye, Sethuraman Panchanathan, Wei Fan, Ian N. Davidson. 2011

57 citations recorded by the source corpus.

Subfields: Biomedical Engineering, Experimental and Cognitive Psychology, Human-Computer Interaction

How researchers see this paper

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

In Biomedical Engineering

1. Gabriella Csurka
2. Jieping Ye
3. Bernhard Schölkopf
4. Jeff Schneider
5. Sebastian Ruder
6. Mingming Gong
7. Kun Zhang
8. Yu Zhang

In Experimental and Cognitive Psychology

1. Gabriella Csurka
2. Jieping Ye
3. Bernhard Schölkopf
4. Sethuraman Panchanathan
5. Qian Sun
6. Rita Chattopadhyay

In Human-Computer Interaction

1. Gabriella Csurka

Citations in the corpus

• Multi-Source Domain Adaptation: A Causal View (2015, background): “2009), and (Chattopadhyay et al. 2011) combine the predictions made by the source hypotheses, with the weights determined in different ways.”
• Domain Adaptation for Visual Applications: A Comprehensive Survey (2017, methodology): “Domain generalization [151, 254, 97, 94, 155], similarly to multi-source DA [31, 70, 120], aims to extract and average knowledge from several related source domains, in order to learn a model for data in a new target domains.”
• Generalization Bounds for Transfer Learning under Model Shift (2015, methodology): “The second baseline optDA is a multi-source domain adaptation algorithm under an optimization framework, as proposed in Chattopadhyay et al. (2011), where the parameters γ A , γ I are set as described in the paper, and θ is chosen using cross-validation on the set { 0 .”
• Knowledge Adaptation: Teaching to Adapt (2017, methodology): “, 2009)), (CP-MDA, (Chattopadhyay et al., 2012)); crossdomain sentiment classification by transferring sentiment along a sentiment graph with hinge loss and logistic loss respectively (SDAMS-SVM and SDAMS-Log) (Wu and Huang, 2016).”
• A Two-Stage Weighting Framework for Multi-Source Domain Adaptation (2011, background): “The second stage computes the weights of multiple sources to reduce the conditional probability differences; the computation is based on the smoothness assumption on the conditional probability distribution of the target domain data [16].”
• On handling negative transfer and imbalanced distributions in multiple source transfer learning (2014, unknown): “Based on this simple philosophy, many methods have been proposed on transfer learning [10] [8] [2] [3] [9] [4] [6] [7] and many successful applications including document classification, WiFi localization, and sentiment classification [12] demonstrate the power of transfer learning.”