论文标题
多类学习者的Pac-Bayesian域适应范围
PAC-Bayesian Domain Adaptation Bounds for Multiclass Learners
论文作者
论文摘要
多类神经网络是现代无监督域适应的常见工具，但是在适应性文献中缺乏针对其非均匀样本复杂性的适当理论描述。为了填补这一空白，我们为多类学习者提出了第一个Pac-Bayesian适应范围。我们还提出了我们考虑的多类分布差异的第一个近似技术，从而促进了界限的实际使用。对于依赖Gibbs预测因子的分歧，我们提出了其他PAC-Bayesian适应界限，以消除对蒙特卡洛效率低下的需求。从经验上讲，我们测试了我们提出的近似技术的功效以及一些新型的设计概念，我们在范围中包括。最后，我们应用界限来分析使用神经网络的常见适应算法。
Multiclass neural networks are a common tool in modern unsupervised domain adaptation, yet an appropriate theoretical description for their non-uniform sample complexity is lacking in the adaptation literature. To fill this gap, we propose the first PAC-Bayesian adaptation bounds for multiclass learners. We facilitate practical use of our bounds by also proposing the first approximation techniques for the multiclass distribution divergences we consider. For divergences dependent on a Gibbs predictor, we propose additional PAC-Bayesian adaptation bounds which remove the need for inefficient Monte-Carlo estimation. Empirically, we test the efficacy of our proposed approximation techniques as well as some novel design-concepts which we include in our bounds. Finally, we apply our bounds to analyze a common adaptation algorithm that uses neural networks.
