Determining mortality risk is important for critical decisions in Intensive Care Units (ICU). The need for machine learning models that provide accurate patient-specific prediction of mortality is well recognized. We present a new algorithm for ICU mortality prediction that is designed to address the problem of imbalance, which occurs, in the context of binary classification, when one of the two classes is significantly under--represented in the data. We take a fundamentally new approach in exploiting the class imbalance through a feature transformation such that the transformed features are easier to classify. Hypothesis testing is used for classification with a test statistic that follows the distribution of the difference of two chi-squared random variables, for which there are no analytic expressions and we derive an accurate approximation. Experiments on a benchmark dataset of 4000 ICU patients show that our algorithm surpasses the best competing methods for mortality prediction.

Automated planning can be used to efficiently recognize goals and plans from partial or full observed action sequences. In this paper, we propose goal recognition heuristics that rely on information from planning landmarks - facts or actions that must occur if a plan is to achieve a goal when starting from some initial state. We develop two such heuristics: the first estimates goal completion by considering the ratio between achieved and extracted landmarks of a candidate goal, while the second takes into account how unique each landmark is among landmarks for all candidate goals. We empirically evaluate these heuristics over both standard goal/plan recognition problems, and a set of very large problems. We show that our heuristics can recognize goals more accurately, and run orders of magnitude faster, than the current state-of-the-art.

We propose a scalable approach to learn video-based question answering (QA): to answer a free-form natural language question about the contents of a video. Our approach automatically harvests a large number of videos and descriptions freely available online. Then, a large number of candidate QA pairs are automatically generated from descriptions rather than manually annotated. Next, we use these candidate QA pairs to train a number of video-based QA methods extended from MN (Sukhbaatar et al. 2015), VQA (Antol et al. 2015), SA (Yao et al. 2015), and SS (Venugopalan et al. 2015). In order to handle non-perfect candidate QA pairs, we propose a self-paced learning procedure to iteratively identify them and mitigate their effects in training. Finally, we evaluate performance on manually generated video-based QA pairs. The results show that our self-paced learning procedure is effective, and the extended SS model outperforms various baselines.

A common approach in positive-unlabeled learning is to train a classification model between labeled and unlabeled data. This strategy is in fact known to give an optimal classifier under mild conditions; however, it results in biased empirical estimates of the classifier performance. In this work, we show that the typically used performance measures such as the receiver operating characteristic curve, or the precision recall curve obtained on such data can be corrected with the knowledge of class priors; i.e., the proportions of the positive and negative examples in the unlabeled data. We extend the results to a noisy setting where some of the examples labeled positive are in fact negative and show that the correction also requires the knowledge of the proportion of noisy examples in the labeled positives. Using state-of-the-art algorithms to estimate the positive class prior and the proportion of noise, we experimentally evaluate two correction approaches and demonstrate their efficacy on real-life data.

While recent advances in computer vision have caused object recognition rates to spike, there is still much room for improvement. In this paper, we develop an algorithm to improve object recognition by integrating human-generated contextual information with vision algorithms. Specifically, we examine how interactive systems such as robots can utilize two types of context information--verbal descriptions of an environment and human-labeled datasets. We propose a re-ranking schema, MultiRank, for object recognition that can efficiently combine such information with the computer vision results. In our experiments, we achieve up to 9.4% and 16.6% accuracy improvements using the oracle and the detected bounding boxes, respectively, over the vision-only recognizers. We conclude that our algorithm has the ability to make a significant impact on object recognition in robotics and beyond.

A challenge for mining large-scale streaming data overlooked by most existing studies on online learning is the skew-distribution of examples over different classes. Many previous works have considered cost-sensitive approaches in an online setting for streaming data, where fixed costs are assigned to different classes, or ad-hoc costs are adapted based on the distribution of data received so far. However, it is not necessary for them to achieve optimal performance in terms of the measures suited for imbalanced data, such as F-measure, area under ROC curve (AUROC), area under precision and recall curve (AUPRC). This work proposes a general framework for online learning with imbalanced streaming data, where examples are coming sequentially and models are updated accordingly on-the-fly. By simultaneously learning multiple classifiers with different cost vectors, the proposed method can be adopted for different target measures for imbalanced data, including F-measure, AUROC and AUPRC. Moreover, we present a rigorous theoretical justification of the proposed framework for the F-measure maximization. Our empirical studies demonstrate the competitive if not better performance of the proposed method compared to previous cost-sensitive and resampling based online learning algorithms and those that are designed for optimizing certain measures.

Sports broadcasters are continuously seeking to make their live coverages of soccer matches more attractive. A recent innovation is the “highlight channel,” which shows the most interesting events from multiple matches played at the same time. However, switching between matches at the right time is challenging in fast-paced sports like soccer, where interesting situations often evolve as quickly as they disappear again. This paper presents the POGBA algorithm for automatically predicting highlights in soccer matches, which is an important task that has not yet been addressed. POGBA leverages spatio-temporal event streams collected during matches to predict the probability that a particular game state will lead to a goal. An empirical evaluation on a real-world dataset shows that POGBA outperforms the baseline algorithms in terms of both precision and recall.

Effective human-aware robots should anticipate their user’s intentions. During hand-eye coordination tasks, gaze often precedes hand motion and can serve as a powerful predictor for intent. However, cooperative tasks where a semi-autonomous robot serves as an extension of the human hand have rarely been studied in the context of hand-eye coordination. We hypothesize that accounting for anticipatory eye movements in addition to the movements of the robot will improve intent estimation. This research compares the application of various machine learning methods to intent prediction from gaze tracking data during robotic hand-eye coordination tasks. We found that with proper feature selection, accuracies exceeding 94% and AUC greater than 91% are achievable with several classification algorithms but that anticipatory gaze data did not improve intent prediction.

The distributions can be different (Storkey and Sugiyama 2006; training and test domains are commonly referred to in the Ben-David and Urner 2012; 2014). Covariate shift is a valid domain adaptation literature as the source and target domains, assumption in some problems, but it can as well be quite respectively. Domain diversity can emerge as a result of the unrealistic for many other domain adaptation tasks where the scarcity of available labeled data from the target domain. It conditional label distributions are not (or, more precisely, not can as well be innate in the problem itself due to, for example, guaranteed to be) identical. The simplification resulting from an ongoing change occurring to the source domain like assuming identical labeling distributions facilitates the quest in cases where the original source domain keeps changing for a tractable learning algorithm, albeit possibly at the cost over time. Domain adaptation aims at finding solutions for of reducing the expressiveness power of the representation, this kind of problem, where the training (source) data are and consequently the accuracy of the resulting hypothesis.

Analyses of text corpora over time can reveal trends in beliefs, interest, and sentiment about a topic. We focus on views expressed about artificial intelligence (AI) in the New York Times over a 30-year period. General interest, awareness, and discussion about AI has waxed and waned since the field was founded in 1956. We present a set of measures that captures levels of engagement, measures of pessimism and optimism, the prevalence of specific hopes and concerns, and topics that are linked to discussions about AI over decades. We find that discussion of AI has increased sharply since 2009, and that these discussions have been consistently more optimistic than pessimistic. However, when we examine specific concerns, we find that worries of loss of control of AI, ethical concerns for AI, and the negative impact of AI on work have grown in recent years. We also find that hopes for AI in healthcare and education have increased over time.