Exploiting Task Relatedness for Mulitple Task Learning

By Shai Ben-David, Reba Schuller Borbely, Rebecca A. Schuller. 2003

496 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computational Mechanics, Optimization & Training

How researchers see this paper

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

In Artificial Intelligence

In Computational Mechanics

In Optimization & Training

Citations in the corpus

An Overview of Multi-Task Learning in Deep Neural Networks (2017, background): “[Ben-David and Schuller, 2003] propose that two tasks are F -related if the data for both tasks can be generated from a ﬁxed probability distribution using a set of transformations F .”
Semi-Cyclic Stochastic Gradient Descent (2019, background): “In order to do so, it is necessary to target a particular way in which tasks are related (Baxter, 2000; Ben-David & Schuller, 2003).”
A Survey on Multi-Task Learning (2017, background): “[196], [197], [198] study the generalization bound of MTL based on the VC dimension by assuming that data distributions of different tasks can be transformed via some known family of functions.”
Undoing the Damage of Dataset Bias (2012, methodology): “A mathematical framework similar to ours is proposed in [11, 12] for multi-task learning, where solutions to multiple tasks are tied through a common weight vector.”
Latent Multi-Task Architecture Learning (2017, background): “ and b) while sluice networks easily ﬁt noise, they are robust across domains in practice. 1 Motivation While there is theory providing guarantees for effectivity of some types of multi-task learning [2, 3], none of these hold for the loosely related tasks to which multi-task learning is most often ”
Efficient Representations for Lifelong Learning and Autoencoding (2014, background): “Other related sample complexity results appear in [8].”
The Benefit of Multitask Representation Learning (2015, methodology): “Despite the considerable success of multitask learning and in particular multitask representation learning there are only few theoretical investigations (Ando and Zhang, 2005; Baxter, 2000; Ben-David and Schuller, 2003).”
Sluice networks: Learning what to share between loosely related tasks (2017, background): “While there is theory providing guarantees for effectivity of some types of multi-task learning [2, 3], none of these hold for the loosely related tasks to which multi-task learning is most often applied in natural language processing (NLP) or computer vision [8, 25].”
Sparse coding for multitask and transfer learning (2012, background): “it was generated, and in the domain of learning to learn, where the dictionary is applied to new tasks of the same environment. Our work combines ideas from sparse coding [25, 26], multitask learning [1, 3, 10, 13, 14] and learning to learn [7, 30]. There is a vast literature on these subjects and t”
Excess risk bounds for multitask learning with trace norm regularization (2012, methodology): “[9] uses a special assumption of task-relatedness to give interesting bounds not on the average, but the maximal risk over the tasks.”
Analysis of Representations for Domain Adaptation (2006, background): “While many authors have analyzed adaptation from multiple sets of labeled training data [3, 5, 7, 13], our theory applies to the setting in which the target domain has no labeled training data, but plentiful unlabeled data exists for both target and source domains.”
Hierarchical Multitask Structured Output Learning for Large-scale Sequence Segmentation (2011, background): “, [5, 8]) have been explored, providing empirical findings [16] and theoretical foundations [3, 4].”
Local Rademacher Complexity-based Learning Guarantees for Multi-Task Learning (2016, background): “Also, in the context of domain adaptation, similar considerations are examined by [41, 43, 42, 21, 66, 44] and [22].”