Anoop R Katti, Christian Reisswig, Cordula Guder, Sebastian Brarda, Steffen Bickel, Johannes Höhne, Jean Baptiste Faddoul. Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing. 2018.

MOTIVATION: As there exists no cure or vaccine for the infection with human immunodeficiency virus (HIV), the standard approach to treating HIV patients is to repeatedly administer different combinations of several antiretroviral drugs. Because of the large number of possible drug combinations, manually finding a successful regimen becomes practically impossible. This presents a major challenge for HIV treatment. The application of machine learning methods for predicting virological responses to potential therapies is a possible approach to solving this problem. However, due to evolving trends in treating HIV patients the available clinical datasets have a highly unbalanced representation, which might negatively affect the usefulness of derived statistical models. RESULTS: This article presents an approach that tackles the problem of predicting virological response to combination therapies by learning a separate logistic regression model for each therapy. The models are fitted by using not only the data from the target therapy but also the information from similar therapies. For this purpose, we introduce and evaluate two different measures of therapy similarity. The models are also able to incorporate phenotypic knowledge on the therapy outcomes through a Gaussian prior. With our approach we balance the uneven therapy representation in the datasets and produce higher quality models for therapies with very few training samples. According to the results from the computational experiments our therapy similarity model performs significantly better than training separate models for each therapy by using solely their examples. Furthermore, the model's performance is as good as an approach that encodes therapy information in the input feature space with the advantage of delivering better results for therapies with very few training samples. AVAILABILITY: Code of the efficient logistic regression is available from http://www.mpi-inf.mpg.de/%7Ejasmina/fastLogistic.zip.

We address the problem of learning classifiers for a large number of tasks. We derive a solution that produces resampling weights which match the pool of all examples to the target distribution of any given task. Our work is motivated by the problem of predicting the outcome of a therapy attempt for a patient who carries an HIV virus with a set of observed genetic properties. Such predictions need to be made for hundreds of possible combinations of drugs, some of which use similar biochemical mechanisms. Multi-task learning enables us to make predictions even for drug combinations with few or no training examples and substantially improves the overall prediction accuracy.

We address the problem of learning classifiers for several related tasks that may differ in their joint distribution of input and output variables. For each task, small - possibly even empty - labeled samples and large unlabeled samples are available. While the unlabeled samples reflect the target distribution, the labeled samples may be biased. We derive a solution that produces resampling weights which match the pool of all examples to the target distribution of any given task. Our work is motivated by the problem of predicting sociodemographic features for users of web portals, based on the content which they have accessed. Here, questionnaires offered to a small portion of each portal's users produce biased samples. Transfer learning enables us to make predictions even for new portals with few or no training data and improves the overall prediction accuracy.

Covariate shift is a situation in supervised learning where training and test inputs follow different distributions even though the functional relation remains unchanged. A common approach to compensating for the bias caused by covariate shift is to reweight the training samples according to importance, which is the ratio of test and training densities. We propose a novel method that allows us to directly estimate the importance from samples without going through the hard task of density estimation. An advantage of the proposed method is that the computation time is nearly independent of the number of test input samples, which is highly beneficial in recent applications with large numbers of unlabeled samples. We demonstrate through experiments that the proposed method is computationally more efficient than existing approaches with comparable accuracy.

We address classification problems for which the training instances are governed by a distribution that is allowed to differ arbitrarily from the test distribution---problems also referred to as classification under covariate shift. We derive a solution that is purely discriminative: neither training nor test distribution are modeled explicitly. We formulate the general problem of learning under covariate shift as an integrated optimization problem. We derive a kernel logistic regression classifier for differing training and test distributions.

Abstract Publication repositories contain an abundance of information about the evolution of scientific research areas. We address the problem of creating a visualization of a research area that describes the flow of topics between papers, quantifies the impact that papers have on each other, and helps to identify key contributions. To this end, we devise a probabilistic topic model that explains the generation of documents; the model incorporates the aspects of topical innovation and topical inheritance via citations. We evaluate the model's ability to predict the strength of influence of citations against manually rated citations.

We study a setting that is motivated by the problem of filtering spam messages for many users. Each user receives messages according to an individual, unknown distribution, reflected only in the unlabeled inbox. The spam filter for a user is required to perform well with respect to this distribution. Labeled messages from publicly available sources can be utilized, but they are governed by a distinct distribution, not adequately representing most inboxes. We devise a method that minimizes a loss function with respect to a user's personal distribution based on the available biased sample. A nonparametric hierarchical Bayesian model furthermore generalizes across users by learning a common prior which is imposed on new email accounts. Empirically, we observe that bias-corrected learning outperforms naive reliance on the assumption of independent and identically distributed data; Dirichlet-enhanced generalization across users outperforms a single ("one size fits all") filter as well as independent filters for all users.

In task 1A of the BioCreAtIvE evaluation, systems had to be devised that recognize words and phrases forming gene or protein names in natural language sentences. We approach this problem by building a word classification system based on a sliding window approach with a Support Vector Machine, combined with a pattern-based post-processing for the recognition of phrases. The performance of such a system crucially depends on the type of features chosen for consideration by the classification method, such as pre- or postfixes, character n-grams, patterns of capitalization, or classification of preceding or following words. We present a systematic approach to evaluate the performance of different feature sets based on recursive feature elimination, RFE. Based on a systematic reduction of the number of features used by the system, we can quantify the impact of different feature sets on the results of the word classification problem. This helps us to identify descriptive features, to learn about the structure of the problem, and to design systems that are faster and easier to understand. We observe that the SVM is robust to redundant features. RFE improves the performance by 0.7%, compared to using the complete set of attributes. Moreover, a performance that is only 2.3% below this maximum can be obtained using fewer than 5% of the features.

We consider clustering problems in which the available attributes can be split into two independent subsets, such that either subset suffices for learning. Example applications of this multi-view setting include clustering of Web pages which have an intrinsic view (the pages themselves) and an extrinsic view (e.g., anchor texts of inbound hyperlinks); multi-view learning has so far been studied in the context of classification. We develop and study partitioning and agglomerative, hierarchical multi-view clustering algorithms for text data. We find empirically that the multi-view versions of k-means and EM greatly improve on their single-view counterparts. By contrast, we obtain negative results for agglomerative hierarchical multi-view clustering. Our analysis explains this surprising phenomenon.