The increase of publicly available bioactivity data in recent years has fueled and catalyzed research in chemogenomics, data mining, and modeling approaches. As a direct result, over the past few years a multitude of different methods have been reported and evaluated, such as target fishing, nearest neighbor similarity-based methods, and Quantitative Structure Activity Relationship (QSAR)-based protocols. However, such studies are typically conducted on different datasets, using different validation strategies, and different metrics. In this study, different methods were compared using one single standardized dataset obtained from ChEMBL, which is made available to the public, using standardized metrics (BEDROC and Matthews Correlation Coefficient). Specifically, the performance of Naïve Bayes, Random Forests, Support Vector Machines, Logistic Regression, and Deep Neural Networks was assessed using QSAR and proteochemometric (PCM) methods. All methods were validated using both a random split validation and a temporal validation, with the latter being a more realistic benchmark of expected prospective execution. Deep Neural Networks are the top performing classifiers, highlighting the added value of Deep Neural Networks over other more conventional methods. Moreover, the best method (‘DNN_PCM’) performed significantly better at almost one standard deviation higher than the mean performance. Furthermore, Multi-task and PCM implementations were shown to improve performance over single task Deep Neural Networks. Conversely, target prediction performed almost two standard deviations under the mean performance. Random Forests, Support Vector Machines, and Logistic Regression performed around mean performance. Finally, using an ensemble of DNNs, alongside additional tuning, enhanced the relative performance by another 27% (compared with unoptimized ‘DNN_PCM’). Here, a standardized set to test and evaluate different machine learning algorithms in the context of multi-task learning is offered by providing the data and the protocols.Graphical Abstract. Electronic supplementary materialThe online version of this article (doi:10.1186/s13321-017-0232-0) contains supplementary material, which is available to authorized users.

The increase of publicly available bioactivity data in recent years has fueled and catalyzed research in chemogenomics, data mining, and modeling approaches. As a direct result, over the past few years a multitude of different methods have been reported and evaluated, such as target fishing, nearest neighbor similarity-based methods, and Quantitative Structure Activity Relationship (QSAR)-based protocols. However, such studies are typically conducted on different datasets, using different validation strategies, and different metrics.In this study, different methods were compared using one single standardized dataset obtained from ChEMBL, which is made available to the public, using standardized metrics (BEDROC and Matthews Correlation Coefficient). Specifically, the performance of Naive Bayes, Random Forests, Support Vector Machines, Logistic Regression, and Deep Neural Networks was assessed using QSAR and proteochemometric (PCM) methods. All methods were validated using both a random split validation and a temporal validation, with the latter being a more realistic benchmark of expected prospective execution.Deep Neural Networks are the top performing classifiers, highlighting the added value of Deep Neural Networks over other more conventional methods. Moreover, the best method (‘DNN_PCM’) performed significantly better at almost one standard deviation higher than the mean performance. Furthermore, Multi task and PCM implementations were shown to improve performance over single task Deep Neural Networks. Conversely, target prediction performed almost two standard deviations under the mean performance. Random Forests, Support Vector Machines, and Logistic Regression performed around mean performance. Finally, using an ensemble of DNNs, alongside additional tuning, enhanced the relative performance by another 27% (compared with unoptimized DNN_PCM).Here, a standardized set to test and evaluate different machine learning algorithms in the context of multitask learning is offered by providing the data and the protocols.

This paper presents WODAN2.0, a workflow using Deep Learning for the automated detection of multiple archaeological object classes in LiDAR data from the Netherlands. WODAN2.0 is developed to rapidly and systematically map archaeology in large and complex datasets. To investigate its practical value, a large, random test dataset—next to a small, non-random dataset—was developed, which better represents the real-world situation of scarce archaeological objects in different types of complex terrain. To reduce the number of false positives caused by specific regions in the research area, a novel approach has been developed and implemented called Location-Based Ranking. Experiments show that WODAN2.0 has a performance of circa 70% for barrows and Celtic fields on the small, non-random testing dataset, while the performance on the large, random testing dataset is lower: circa 50% for barrows, circa 46% for Celtic fields, and circa 18% for charcoal kilns. The results show that the introduction of Location-Based Ranking and bagging leads to an improvement in performance varying between 17% and 35%. However, WODAN2.0 does not reach or exceed general human performance, when compared to the results of a citizen science project conducted in the same research area.

It has been recently demonstrated that the classical EM algorithm for learning Gaussian mixture models can be successfully implemented in a decentralized manner by resorting to gossip-based randomized distributed protocols. In this paper we describe a gossip-based implementation of an alternative algorithm for learning Gaussian mixtures in which components are added to the mixture one after another. Our new Greedy Gossip-based Gaussian mixture learning algorithm uses gossip-based parallel search, starting from multiple initial guesses, for finding good components to add to the mixture in each component allocation step. It can be executed on massive networks of small computing devices, converging to a solution exponentially faster than its centralized version, while reaching the same quality of generated models.

Abstract. The research reported in this paper is concerned with assessing the usefulness of reinforcment learning (RL) for on-line calibration of parameters in evolutionary algorithms (EA). We are running an RL procedure and the EA simultaneously and the RL is changing the EA parameters on-the-fly. We evaluate this approach experimentally on a range of fitness landscapes with varying degrees of ruggedness. The results show that EA calibrated by the RL-based approach outperforms a benchmark EA.