Motivated by online advertising auctions, we consider repeated Vickrey auctions where goods of unknown value are sold sequentially and bidders only learn (potentially noisy) information about a good's value once it is purchased. We adopt an online learning approach with bandit feedback to model this problem and derive bidding strategies for two models: stochastic and adversarial. In the stochastic model, the observed values of the goods are random variables centered around the true value of the good. In this case, logarithmic regret is achievable when competing against well behaved adversaries. In the adversarial model, the goods need not be identical and we simply compare our performance against that of the best fixed bid in hindsight. We show that sublinear regret is also achievable in this case and prove matching minimax lower bounds. To our knowledge, this is the first complete set of strategies for bidders participating in auctions of this type.

Intelligent systems rely on pattern recognition and signature-based approaches for a wide range of sensors enhancing situational awareness. For example, autonomous systems depend on environmental sensors to perform their tasks and secure systems depend on anomaly detection methods. The availability of large amount of data requires the processing of data in a “streaming” fashion with online algorithms. Yet, just as online learning can enhance adaptability to a non-stationary environment, it introduces vulnerabilities that can be manipulated by adversaries to achieve their goals while evading detection. Although human intelligence might have evolved from social interactions, machine intelligence has evolved as a human intelligence artifact and been kept isolated to avoid ethical dilemmas. As our adversaries become sophisticated, it might be time to revisit this question and examine how we can combine online learning and reasoning leading to the science of deceptive and counter-deceptive machines.

We consider a sequential learning problem with Gaussian payoffs and side information: after selecting an action $i$, the learner receives information about the payoff of every action $j$ in the form of Gaussian observations whose mean is the same as the mean payoff, but the variance depends on the pair $(i,j)$ (and may be infinite). The setup allows a more refined information transfer from one action to another than previous partial monitoring setups, including the recently introduced graph-structured feedback case. For the first time in the literature, we provide non-asymptotic problem-dependent lower bounds on the regret of any algorithm, which recover existing asymptotic problem-dependent lower bounds and finite-time minimax lower bounds available in the literature. We also provide algorithms that achieve the problem-dependent lower bound (up to some universal constant factor) or the minimax lower bounds (up to logarithmic factors).

We propose a general framework for studying adaptive regret bounds in the online learning framework, including model selection bounds and data-dependent bounds. Given a data- or model-dependent bound we ask, "Does there exist some algorithm achieving this bound?" We show that modifications to recently introduced sequential complexity measures can be used to answer this question by providing sufficient conditions under which adaptive rates can be achieved. In particular each adaptive rate induces a set of so-called offset complexity measures, and obtaining small upper bounds on these quantities is sufficient to demonstrate achievability. A cornerstone of our analysis technique is the use of one-sided tail inequalities to bound suprema of offset random processes. Our framework recovers and improves a wide variety of adaptive bounds including quantile bounds, second-order data-dependent bounds, and small loss bounds. In addition we derive a new type of adaptive bound for online linear optimization based on the spectral norm, as well as a new online PAC-Bayes theorem that holds for countably infinite sets.

This paper revisits the problem of learning a k-CNF Boolean function from examples, for fixed k, in the context of online learning under the logarithmic loss. We give a Bayesian interpretation to one of Valiant’s classic PAC learning algorithms, which we then build upon to derive three efficient, online, probabilistic, supervised learning algorithms for predicting the output of an unknown k-CNF Boolean function. We analyze the loss of our methods, and show that the cumulative log-loss can be upper bounded by a polynomial function of the size of each example.

A major challenge in Content-Based Image Retrieval (CBIR) is to bridge the semantic gap between low-level image contents and high-level semantic concepts. Although researchers have investigated a variety of retrieval techniques using different types of features and distance functions, no single best retrieval solution can fully tackle this challenge. In a real-world CBIR task, it is often highly desired to combine multiple types of different feature representations and diverse distance measures in order to close the semantic gap. In this paper, we investigate a new framework of learning to rank for CBIR, which aims to seek the optimal combination of different retrieval schemes by learning from large-scale training data in CBIR. We first formulate the problem formally as a learning to rank task, which can be solved in general by applying the existing batch learning to rank algorithms from text information retrieval (IR). To further address the scalability towards large-scale online CBIR applications, we present a family of online learning to rank algorithms, which are significantly more efficient and scalable than classical batch algorithms for large-scale online CBIR. Finally, we conduct an extensive set of experiments, in which encouraging results show that our technique is effective, scalable and promising for large-scale CBIR.

In Massively Open Online Courses (MOOCs) TA resources are limited; most MOOCs use peer assessments to grade assignments. Students have to divide up their time between working on their own homework and grading others. If there is no risk of being caught and penalized, students have no reason to spend any time grading others Course staff want to incentivize students to balance their time between course work and peer grading. They may do so by auditing students, ensuring that they perform grading correctly. One would not want students to invest too much time on peer grading, as this would result in poor course performance. We present the first model of strategic auditing in peer grading, modeling the student's choice of effort in response to a grader's audit levels as a Stackelberg game with multiple followers. We demonstrate that computing the equilibrium for this game is computationally hard. We then provide a PTAS in order to compute an approximate solution to the problem of allocating audit levels. However, we show that this allocation does not necessarily maximize social welfare; in fact, there exist settings where course auditor utility is arbitrarily far from optimal under an approximately optimal allocation. To circumvent this issue, we present a natural condition that guarantees that approximately optimal TA allocations guarantee approximately optimal welfare for the course auditors.

Cognitive modelling can discover the latent characteristics of examinees for predicting their performance (i.e. scores) on each problem. As cognitive modelling is important for numerous applications, e.g. personalized remedy recommendation, some solutions have been designed in the literature. However, the problem of extracting information from both objective and subjective problems to get more precise and interpretable cognitive analysis is still underexplored. To this end, we propose a fuzzy cognitive diagnosis framework (FuzzyCDF) for examinees' cognitive modelling with both objective and subjective problems. Specifically, to handle the partially correct responses on subjective problems, we first fuzzify the skill proficiency of examinees. Then, we combine fuzzy set theory and educational hypotheses to model the examinees' mastery on the problems. Further, we simulate the generation of examination scores by considering both slip and guess factors. Extensive experiments on three real-world datasets prove that FuzzyCDF can predict examinee performance more effectively, and the output of FuzzyCDF is also interpretative.

Parallel Test Paper Generation ( k -TPG) is a biobjective distributed resource allocation problem, which aims to generate multiple similarly optimal test papers automatically according to multiple user-specified criteria.Generating high-quality parallel test papers is challenging due to its NP-hardness in maximizing the collective objective functions.In this paper, we propose a Collective Biobjective Optimization (CBO) algorithm for solving k -TPG. CBO is a multi-step greedy-based approximation algorithm, which exploits the submodular property for biobjective optimization of k -TPG.Experiment results have shown that CBO has drastically outperformed the current techniques in terms of paper quality and runtime efficiency.

This paper introduces a new probabilistic model for online learning which dynamically incorporates information from stochastic gradients of an arbitrary loss function. Similar to probabilistic filtering, the model maintains a Gaussian belief over the optimal weight parameters. Unlike traditional Bayesian updates, the model incorporates a small number of gradient evaluations at locations chosen using Thompson sampling, making it computationally tractable. The belief is then transformed via a linear flow field which optimally updates the belief distribution using rules derived from information theoretic principles. Several versions of the algorithm are shown using different constraints on the flow field and compared with conventional online learning algorithms. Results are given for several classification tasks including logistic regression and multilayer neural networks.