A major assumption in many machine learning and data mining algorithms is that the training and future data must be in the same feature space and have the same distribution. However, in many real-world applications, this assumption may not hold. For example, we sometimes have a classification task in one domain of interest, but we only have sufficient training data in another domain of interest, where the latter data may be in a different feature space or follow a different data distribution. In such cases, knowledge transfer, if done successfully, would greatly improve the performance of learning by avoiding much expensive data labeling efforts. In recent years, transfer learning has emerged as a new learning framework to address this problem. This survey focuses on categorizing and reviewing the current progress on transfer learning for classification, regression and clustering problems. In this survey, we discuss the relationship between transfer learning and other related machine learning techniques such as domain adaptation, multitask learning and sample selection bias, as well as co-variate shift. We also explore some potential future issues in transfer learning research.

The goal of transfer learning is to use the knowledge acquired in a set of source tasks to improve performance in a related but previously unseen target task. In this paper, we present a multilayered architecture named CAse-Based Reinforcement Learner (CARL). It uses a novel combination of Case-Based Reasoning (CBR) and Reinforcement Learning (RL) to achieve transfer while playing against the Game AI across a variety of scenarios in MadRTS TM, a commercial Real Time Strategy game. Our experiments demonstrate that CARL not only performs well on individual tasks but also exhibits significant performance gains when allowed to transfer knowledge from previous

Abstract. We introduce a problem called relevant subtask learning, a variant of multi-task learning. The goal is to build a classifier for a taskof-interest having too little data. We additionally have data for other tasks but only some are relevant, meaning that they contain samples classified in the same way as in the task-of-interest. The problem is how to utilize this “background data ” to improve the classifier in the task-ofinterest. We show how to solve the problem for logistic regression classifiers, and demonstrate that the solution works better than a comparable multi-task learning model. The key is to assume that data of all tasks are mixtures of relevant and irrelevant samples, and model the irrelevant part with a sufficiently flexible model such that it does not distort the model of relevant data. Key words: multi-task learning, partially relevant data, relevant subtask learning, transfer learning 1

The sparsity problem in collaborative filtering (CF) is a major bottleneck for most CF methods. In this paper, we consider a novel approach for alleviating the sparsity problem in CF by transferring useritem rating patterns from a dense auxiliary rating matrix in other domains (e.g., a popular movie rating website) to a sparse rating matrix in a target domain (e.g., a new book rating website). We do not require that the users and items in the two domains be identical or even overlap. Based on the limited ratings in the target matrix, we establish a bridge between the two rating matrices at a clusterlevel of user-item rating patterns in order to transfer more useful knowledge from the auxiliary task domain. We first compress the ratings in the auxiliary rating matrix into an informative and yet compact cluster-level rating pattern representation referred to as a codebook. Then, we propose an efficient algorithm for reconstructing the target rating matrix by expanding the codebook. We perform extensive empirical tests to show that our method is effective in addressing the data sparsity problem by transferring the useful knowledge from the auxiliary tasks, as compared to many state-of-the-art CF methods. 1

Learning object categories from small samples is a challenging problem, where machine learning tools can in general provide very few guarantees. Exploiting prior knowledge may be useful to reproduce the human capability of recognizing objects even from only one single view. This paper presents an SVM-based model adaptation algorithm able to select and weight appropriately prior knowledge coming from different categories. The method relies on the solution of a convex optimization problem which ensures to have the minimal leave-one-out error on the training set. Experiments on a subset of the Caltech-256 database show that the proposed method produces better results than both choosing one single prior model, and transferring from all previous experience in a flat uninformative way. 1.

Abstract. Transfer learning is the improvement of learning in a new task through the transfer of knowledge from a related task that has already been learned. While most machine learning algorithms are designed to address single tasks, the development of algorithms that facilitate transfer learning is a topic of ongoing interest in the machine-learning community. This chapter provides an introduction to the goals, formulations, and challenges of transfer learning. It surveys current research in this area, giving an overview of the state of the art and outlining the open problems. The survey covers transfer in both inductive learning and reinforcement learning, and discusses the issues of negative transfer and task mapping in depth.

Abstract. Most existing transfer learning techniques are limited to problems of knowledge transfer across tasks sharing the same set of class labels. In this paper, however, we relax this constraint and propose a spectral-based solution that aims at unveiling the intrinsic structure of the data and generating a partition of the target data, by transferring the eigenspace that well separates the source data. Furthermore, a clusteringbased KL divergence is proposed to automatically adjust how much to transfer. We evaluate the proposed model on text and image datasets where class categories of the source and target data are explicitly different, e.g., 3-classes transfer to 2-classes, and show that the proposed approach improves other baselines by an average of 10 % in accuracy. The source code and datasets are available from the authors. 1

Inductive transfer is applying knowledge learned on one set of tasks to improve the performance of learning a new task. Inductive transfer is being applied in improving the generalization performance on a classification task using the models learned on some related tasks. In this paper, we show a method of making inductive transfer for text classification more effective using Wikipedia. We map the text documents of the different tasks to a feature space created using Wikipedia, thereby providing some background knowledge of the contents of the documents. It has been observed here that when the classifiers are built using the features generated from Wikipedia they become more effective in transferring knowledge. An evaluation on the daily classification task on the Reuters RCV1 corpus shows that our method can significantly improve the performance of inductive transfer. Our method was also able to successfully overcome a major obstacle observed in a recent work on a similar setting. 1.

Abstract. Domain adaptation (DA) using labeled data from related source domains comes in handy when the labeled patterns of a target domain are scarce. Nevertheless, it is worth noting that when the predictive distribution P (y|x) of the domains differs, which establishes Negative Transfer [19], DA approaches generally fail to perform well. Taking this cue, the Predictive Distribution Matching SVM (PDM-SVM) is proposed to learn a robust classifier in the target domain (referred to as the target classifier) by leveraging the labeled data from only the relevant regions of multiple sources. In particular, a k-nearest neighbor graph is iteratively constructed to identify the regions of relevant source labeled data where the predictive distribution maximally aligns with that of the target data. Predictive distribution matching regularization is then introduced to leverage these relevant source labeled data for training the target classifier. In addition, progressive transduction is adopted to infer the label of target unlabeled data for estimating the predictive distribution of the target domain. Finally, extensive experiments are conducted to illustrate the impact of Negative Transfer on several existing state-of-the-art DA methods, and demonstrate the improved performance efficacy of our proposed PDM-SVM on the commonly used multi-domain Sentiment and Reuters datasets.

In recent years, knowledge transfer algorithms have become one of most the active research areas in learning visual concepts. Most of the existing learning algorithms focuses on leveraging the knowledge transfer process which is