Corpus ID: 211258879

The Sample Complexity of Meta Sparse Regression

@article{Wang2020TheSC,
  title={The Sample Complexity of Meta Sparse Regression},
  author={Zhan-Yu Wang and Jean Honorio},
  journal={ArXiv},
  year={2020},
  volume={abs/2002.09587}
}
  • Zhan-Yu Wang, Jean Honorio
  • Published 2020
  • Computer Science, Mathematics
  • ArXiv
  • This paper addresses the meta-learning problem in sparse linear regression with infinite tasks. We assume that the learner can access several similar tasks. The goal of the learner is to transfer knowledge from the prior tasks to a similar but novel task. For p parameters, size of the support set k , and l samples per task, we show that T \in O (( k log(p) ) /l ) tasks are sufficient in order to recover the common support of all tasks. With the recovered support, we can greatly reduce the… CONTINUE READING

    Topics from this paper.

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 28 REFERENCES
    A Dirty Model for Multi-task Learning
    313
    Taking Advantage of Sparsity in Multi-Task Learning
    234
    On the Convergence Theory of Gradient-Based Model-Agnostic Meta-Learning Algorithms
    30
    Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
    2160
    Online Meta-Learning
    79
    Optimization as a Model for Few-Shot Learning
    1136
    Prototypical Networks for Few-shot Learning
    1410
    Few-shot Learning: A Survey
    56