Self-training based semi-supervised learning algorithms have enabled the learning of highly accurate deep neural networks, using only a fraction of labeled data. However, the majority of work on self-training has focused on the objective of improving accuracy, whereas practical machine learning systems can have complex goals (e.g. maximizing the minimum of recall across classes, etc.) that are non-decomposable in nature. In this work, we introduce the Cost-Sensitive Self-Training (CSST) framework which generalizes the self-training-based methods for optimizing non-decomposable metrics. We prove that our framework can better optimize the desired non-decomposable metric utilizing unlabeled data, under similar data distribution assumptions made for the analysis of self-training. Using the proposed CSST framework, we obtain practical self-training methods (for both vision and NLP tasks) for optimizing different non-decomposable metrics using deep neural networks. Our results demonstrate that CSST achieves an improvement over the state-of-the-art in majority of the cases across datasets and objectives.
The business objectives of recommenders, such as increasing sales, are aligned with the causal effect of recommendations. Previous recommenders targeting for the causal effect employ the inverse propensity scoring (IPS) in causal inference. However, IPS is prone to suffer from high variance. The matching estimator is another representative method in causal inference field. It does not use propensity and hence free from the above variance problem. In this work, we unify traditional neighborhood recommendation methods with the matching estimator, and develop robust ranking methods for the causal effect of recommendations. Our experiments demonstrate that the proposed methods outperform various baselines in ranking metrics for the causal effect. The results suggest that the proposed methods can achieve more sales and user engagement than previous recommenders.
The RKHS bandit problem (also called kernelized multi-armed bandit problem) is an online optimization problem of non-linear functions with noisy feedbacks. Most of the existing methods for the problem have sub-linear regret guarantee at the cost of high computational complexity. For example, IGP-UCB requires at least quadratic time in the number of observed samples at each round. In this paper, using deep results provided by the approximation theory, we approximately reduce the problem to the well-studied linear bandit problem of an appropriate dimension. Then, we propose several algorithms and prove that they achieve comparable regret guarantee to the existing methods (GP-UCB, IGP-UCB) with less computational complexity. Specifically, our proposed methods require polylogarithmic time to select an arm at each round for kernels with "infinite smoothness" (e.g. the rational quadratic and squared exponential kernels). Furthermore, we empirically show our proposed method has comparable regret to the existing method and its running time is much shorter.
Increasing users' positive interactions, such as purchases or clicks, is an important objective of recommender systems. Recommenders typically aim to select items that users will interact with. If the recommended items are purchased, an increase in sales is expected. However, the items could have been purchased even without recommendation. Thus, we want to recommend items that results in purchases caused by recommendation. This can be formulated as a ranking problem in terms of the causal effect. Despite its importance, this problem has not been well explored in the related research. It is challenging because the ground truth of causal effect is unobservable, and estimating the causal effect is prone to the bias arising from currently deployed recommenders. This paper proposes an unbiased learning framework for the causal effect of recommendation. Based on the inverse propensity scoring technique, the proposed framework first constructs unbiased estimators for ranking metrics. Then, it conducts empirical risk minimization on the estimators with propensity capping, which reduces variance under finite training samples. Based on the framework, we develop an unbiased learning method for the causal effect extension of a ranking metric. We theoretically analyze the unbiasedness of the proposed method and empirically demonstrate that the proposed method outperforms other biased learning methods in various settings.
The linear submodular bandit problem was proposed to simultaneously address diversified retrieval and online learning in a recommender system. If there is no uncertainty, this problem is equivalent to a submodular maximization problem under a cardinality constraint. However, in some situations, recommendation lists should satisfy additional constraints such as budget constraints, other than a cardinality constraint. Thus, motivated by diversified retrieval considering budget constraints, we introduce a submodular bandit problem under the intersection of $l$ knapsacks and a $k$-system constraint. Here $k$-system constraints form a very general class of constraints including cardinality constraints and the intersection of $k$ matroid constraints. To solve this problem, we propose a non-greedy algorithm that adaptively focuses on a standard or modified upper-confidence bound. We provide a high-probability upper bound of an approximation regret, where the approximation ratio matches that of a fast offline algorithm. Moreover, we perform experiments under various combinations of constraints using a synthetic and two real-world datasets and demonstrate that our proposed methods outperform the existing baselines.