Transfer learning is a machine learning paradigm where knowledge from one problem is utilized to solve a new but related problem. While conceivable that knowledge from one task could be useful for solving a related task, if not executed properly, transfer learning algorithms can impair the learning performance instead of improving it -- commonly known as negative transfer. In this paper, we study transfer learning from a Bayesian perspective, where a parametric statistical model is used. Specifically, we study three variants of transfer learning problems, instantaneous, online, and time-variant transfer learning. For each problem, we define an appropriate objective function, and provide either exact expressions or upper bounds on the learning performance using information-theoretic quantities, which allow simple and explicit characterizations when the sample size becomes large. Furthermore, examples show that the derived bounds are accurate even for small sample sizes. The obtained bounds give valuable insights into the effect of prior knowledge for transfer learning, at least with respect to our Bayesian formulation of the transfer learning problem. In particular, we formally characterize the conditions under which negative transfer occurs. Lastly, we devise two (online) transfer learning algorithms that are amenable to practical implementations, one of which does not require the parametric assumption. We demonstrate the effectiveness of our algorithms with real data sets, focusing primarily on when the source and target data have strong similarities.
翻译:转移学习是一种机器学习模式,从一个问题获得的知识被用于解决一个新但相关的问题。虽然可以想象,从一个任务获得的知识,如果执行不当,可能有助于解决相关任务,但转移学习算法会损害学习绩效,而不是改进学习绩效 -- -- 通常称为负转移。在本文中,我们从巴伊西亚的角度研究转移学习,使用的是一个参数统计模型。具体地说,我们研究三个转移学习问题的变式,即即即时、在线和时间变化性转移学习。对于每一个问题,我们确定一个适当的客观功能,并且用信息理论数量提供学习业绩的确切表达或上限值,在样本规模大时允许简单和明确的定性。此外,例子表明,即使对小样本规模而言,衍生的界限也是准确的。我们从巴伊西亚角度研究转移学习知识的影响,至少是我们对转移学习问题的巴伊西亚语公式。我们正式描述进行负转移的条件。最后,我们设计了两个(在线)强的学习算法,在样本规模大时,我们只能将数据与实际执行的对比性假设联系起来,一个数据源。我们需要真正的数据,我们以实际的精确的对比。