•A feature-based transfer learning approach for remaining useful life prediction.•Focus on generalizing models from simple source domains to complex target domains.•Propose distance-to-peers as a ...feature that is transferable across different domains.•The method outperforms TCA, CORAL, SCL, and a non-transfer approach on CMAPSS data.
The traditional paradigm for developing machine prognostics usually relies on generalization from data acquired in experiments under controlled conditions prior to deployment of the equipment. Detecting or predicting failures and estimating machine health in this way assumes that future field data will have a very similar distribution to the experiment data. However, many complex machines operate under dynamic environmental conditions and are used in many different ways. This makes collecting comprehensive data very challenging, and the assumption that pre-deployment data and post-deployment data follow very similar distributions is unlikely to hold.
In this work, we present a feature-representation based transfer learning (TL) method for predicting Remaining Useful Life (RUL) of equipment, under scenarios that samples with previously unseen conditions are presented in the target domain and the labels are available only for the source domain, but not the target domain. This setting corresponds to generalizing from a limited number of run-to-failure experiments performed prior to deployment into making prognostics with data coming from deployed equipment that is being used under multiple new operating conditions and experiencing previously unseen faults. We employ a deviation detection method, Consensus Self-Organizing Models (COSMO), to create transferable features for building the RUL regression model. These features capture how different a particular equipment is in comparison to its peers.
The efficiency of the proposed TL method is demonstrated using the NASA Turbofan Engine Degradation Simulation Data Set. Models using the COSMO transferable features show better performance than other methods on predicting RUL when the target domain is more complex than the source domain.
Several papers have been published where nonlinear machine learning algorithms, e.g. artificial neural networks, support vector machines and decision trees, have been used to model the specificity of ...the HIV-1 protease and extract specificity rules. We show that the dataset used in these studies is linearly separable and that it is a misuse of nonlinear classifiers to apply them to this problem. The best solution on this dataset is achieved using a linear classifier like the simple perceptron or the linear support vector machine, and it is straightforward to extract rules from these linear models. We identify key residues in peptides that are efficiently cleaved by the HIV-1 protease and list the most prominent rules, relating them to experimental results for the HIV-1 protease. Motivation: Understanding HIV-1 protease specificity is important when designing HIV inhibitors and several different machine learning algorithms have been applied to the problem. However, little progress has been made in understanding the specificity because nonlinear and overly complex models have been used. Results: We show that the problem is much easier than what has previously been reported and that linear classifiers like the simple perceptron or linear support vector machines are at least as good predictors as nonlinear algorithms. We also show how sets of specificity rules can be generated from the resulting linear classifiers. Availability: The datasets used are available at http://www.hh.se/staff/bioinf/
Understanding the substrate specificity of human immunodeficiency virus (HIV)-1 protease is important when designing effective HIV-1 protease inhibitors. Furthermore, characterizing and predicting ...the cleavage profile of HIV-1 protease is essential to generate and test hypotheses of how HIV-1 affects proteins of the human host. Currently available tools for predicting cleavage by HIV-1 protease can be improved.
The linear support vector machine with orthogonal encoding is shown to be the best predictor for HIV-1 protease cleavage. It is considerably better than current publicly available predictor services. It is also found that schemes using physicochemical properties do not improve over the standard orthogonal encoding scheme. Some issues with the currently available data are discussed.
The datasets used, which are the most important part, are available at the UCI Machine Learning Repository. The tools used are all standard and easily available.
thorsteinn.rognvaldsson@hh.se.
Recent advances in artificial intelligence and deep machine learning have created a step change in how to measure human development indicators, in particular asset-based poverty. The combination of ...satellite imagery and deep machine learning now has the capability to estimate some types of poverty at a level close to what is achieved with traditional household surveys. An increasingly important issue beyond static estimations is whether this technology can contribute to scientific discovery and, consequently, new knowledge in the poverty and welfare domain. A foundation for achieving scientific insights is domain knowledge, which in turn translates into explainability and scientific consistency. We perform an integrative literature review focusing on three core elements relevant in this context—transparency, interpretability, and explainability—and investigate how they relate to the poverty, machine learning, and satellite imagery nexus. Our inclusion criteria for papers are that they cover poverty/wealth prediction, using survey data as the basis for the ground truth poverty/wealth estimates, be applicable to both urban and rural settings, use satellite images as the basis for at least some of the inputs (features), and the method should include deep neural networks. Our review of 32 papers shows that the status of the three core elements of explainable machine learning (transparency, interpretability, and domain knowledge) is varied and does not completely fulfill the requirements set up for scientific insights and discoveries. We argue that explainability is essential to support wider dissemination and acceptance of this research in the development community and that explainability means more than just interpretability.
Research has shown that certain aspects of human poverty and welfare can be measured with deep machine learning in combination with satellite imagery. High hopes have been expressed from policy and decision makers, but downstream applications are still rare. We suggest that one obstacle to overcome is the lack of explainability in some parts of the scientific process. Future directions would be to develop methods to understand which features in an image that triggers a certain response and relate that to domain knowledge; is the response and features in accordance with theory within the field(s)? Suggested reading: Roscher, Ribana, et al. "Explainable machine learning for scientific insights and discoveries." IEEE Access 8 (2020):42;200–42216.
Recent advances in artificial intelligence and deep machine learning have created a step change in how to measure human development indicators. An increasingly important issue is whether this technology can contribute to scientific discovery and, consequently, new knowledge in the poverty and welfare domain. We show that the status of the three core elements of explainable machine learning (transparency, interpretability, and domain knowledge) is varied and does not completely fulfill the requirements set up for scientific insights and discoveries.
Methods and results are presented for applying supervised machine learning techniques to the task of predicting the need for repairs of air compressors in commercial trucks and buses. Prediction ...models are derived from logged on-board data that are downloaded during workshop visits and have been collected over three years on a large number of vehicles. A number of issues are identified with the data sources, many of which originate from the fact that the data sources were not designed for data mining. Nevertheless, exploiting this available data is very important for the automotive industry as means to quickly introduce predictive maintenance solutions. It is shown on a large data set from heavy duty trucks in normal operation how this can be done and generate a profit.
Random forest is used as the classifier algorithm, together with two methods for feature selection whose results are compared to a human expert. The machine learning based features outperform the human expert features, which supports the idea to use data mining to improve maintenance operations in this domain.
In this paper, we study the “Multi-Robot Routing problem” with min–max objective (MRR-MM) in detail. It involves the assignment of sequentially ordered tasks to robots such that the maximum cost of ...the slowest robot is minimized. The problem description, the different types of formulations, and the methods used across various research communities are discussed in this paper. We propose a new problem formulation by treating this problem as a permutation matrix. A comparative study is done between three methods: Stochastic simulated annealing, deterministic mean-field annealing, and a heuristic-based graph search method. Each method is investigated in detail with several data sets (simulation and real-world), and the results are analysed and compared with respect to scalability, computational complexity, optimality, and its application to real-world scenarios. The paper shows that the heuristic method produces results very quickly with good scalability. However, the solution quality is sub-optimal. On the other hand, when optimal or near-optimal results are required with considerable computational resources, the simulated annealing method proves to be more efficient. However, the results show that the optimal choice of algorithm depends on the dataset size and the available computational budget. The contribution of the paper is three-fold: We study the MRR-MM problem in detail across various research communities. This study also shows the lack of inter-research terminology that has led to different names for the same problem. Secondly, formulating the task allocation problem as a permutation matrix formulation has opened up new approaches to solve this problem. Thirdly, we applied our problem formulation to three different methods and conducted a detailed comparative study using real-world and simulation data.
A set of new algorithms and software tools for automatic protein identification using peptide mass fingerprinting is presented. The software is automatic, fast and modular to suit different ...laboratory needs, and it can be operated either via a Java user interface or called from within scripts. The software modules do peak extraction, peak filtering and protein database matching, and communicate via XML. Individual modules can therefore easily be replaced with other software if desired, and all intermediate results are available to the user. The algorithms are designed to operate without human intervention and contain several novel approaches. The performance and capabilities of the software is illustrated on spectra from different mass spectrometer manufacturers, and the factors influencing successful identification are discussed and quantified. Motivation: Protein identification with mass spectrometric methods is a key step in modern proteomics studies. Some tools are available today for doing different steps in the analysis. Only a few commercial systems integrate all the steps in the analysis, often for only one vendor's hardware, and the details of these systems are not public. Results: A complete system for doing protein identification with peptide mass fingerprints is presented, including everything from peak picking to matching the database protein. The details of the different algorithms are disclosed so that academic researchers can have full control of their tools. Availability: The described software tools are available from the Halmstad University website www.hh.se/staff/bioinf/ Supplementary information: Details of the algorithms are described in supporting information available from the Halmstad University website www.hh.se/staff/bioinf/
Several groups have proposed that genotypic determinants in gag and the gp41 cytoplasmic domain (gp41-CD) reduce protease inhibitor (PI) susceptibility without PI-resistance mutations in protease. ...However, no gag and gp41-CD mutations definitively responsible for reduced PI susceptibility have been identified in individuals with virological failure (VF) while receiving a boosted PI (PI/r)-containing regimen. To identify gag and gp41 mutations under selective PI pressure, we sequenced gag and/or gp41 in 61 individuals with VF on a PI/r (n = 40) or NNRTI (n = 20) containing regimen. We quantified nonsynonymous and synonymous changes in both genes and identified sites exhibiting signal for directional or diversifying selection. We also used published gag and gp41 polymorphism data to highlight mutations displaying a high selection index, defined as changing from a conserved to an uncommon amino acid. Many amino acid mutations developed in gag and in gp41-CD in both the PI- and NNRTI-treated groups. However, in neither gene, were there discernable differences between the two groups in overall numbers of mutations, mutations displaying evidence of diversifying or directional selection, or mutations with a high selection index. If gag and/or gp41 encode PI-resistance mutations, they may not be confined to consistent mutations at a few sites.
Vehicle utilization analysis is an essential tool for manufacturers to understand customer needs, improve equipment uptime, and to collect information for future vehicle and service development. ...Typically today, this behavioral modeling is done on high-resolution time-resolved data with features such as GPS position and fuel consumption. However, high-resolution data is costly to transfer and sensitive from a privacy perspective. Therefore, such data is typically only collected when the customer pays for extra services relying on that data. This motivated us to develop a multi-task ensemble approach to transfer knowledge from the high-resolution data and enable vehicle behavior prediction from low-resolution but high dimensional data that is aggregated over time in the vehicles.
This study proposes a multi-task snapshot-stacked ensemble (MTSSE) deep neural network for vehicle behavior prediction by considering vehicles’ low-resolution operational life records. The multi-task ensemble approach utilizes the measurements to map the low-frequency vehicle usage to the vehicle behaviors defined from the high-resolution time-resolved data. Two data sources are integrated and used: high-resolution data called Dynafleet, and low-resolution so-called Logged Vehicle Data (LVD). The experimental results demonstrate the proposed approach’s effectiveness in predicting the vehicle behavior from low frequency data. With the suggested multi-task snapshot-stacked ensemble deep network, it is shown how low-resolution sensor data can highly contribute to predicting multiple vehicle behaviors simultaneously while using only one single training process.
•A multi-task snapshot-ensemble approach for vehicle behavior modeling is proposed.•The approach relies on low-resolution data to predict long term vehicle behavior.•Vehicle behavior can positively and negatively affect the prediction performance.•Extract task transference to find the tasks that should be trained together.