Search for: All records

Autoencoders represent a significant category of deep learning models and are widely utilized for dimensionality reduction. However, standard Autoencoders are complicated architectures that normally have several layers and many hyper-parameters that require tuning. In this paper, we introduce a new type of autoencoder that we call dynamical system autoencoder (DSAE). Similar to classic autoencoders, DSAEs can effectively handle dimensionality reduction and denoising tasks, and they demonstrate strong performance in several benchmark tasks. However, DSAEs, in some sense, have a more flexible architecture than standard AEs. In particular, in this paper we study simple DSAEs that only have a single layer. In addition, DSAEs provide several theoretical and practical advantages arising from their implementation as iterative maps, which have been well studied over several decades. Beyond the inherent simplicity of DSAEs, we also demonstrate how to use sparse matrices to reduce the number of parameters for DSAEs without sacrificing the performance of our methods. Our simulation studies indicate that DSAEs achieved better performance than the classic autoencoders when the encoding dimension or training sample size was small. Additionally, we illustrate how to use DSAEs, and denoising autoencoders in general. to nerform sunervised learning tasks. 

The accurate detection of chemical agents promotes many national security and public safety goals, and robust chemical detection methods can prevent disasters and support effective response to incidents. Mass spectrometry is an important tool in detecting and identifying chemical agents. However, there are high costs and logistical challenges associated with acquiring sufficient lab-generated mass spectrometry data for training machine learning algorithms, including skilled personnel, sample preparation and analysis required for data generation. These high costs of mass spectrometry data collection hinder the development of machine learning and deep learning models to detect and identify chemical agents. Accordingly, the primary objective of our research is to create a mass spectrometry data generation model whose output (synthetic mass spectrometry data) would enhance the performance of downstream machine learning chemical classification models. Such a synthetic data generation model would reduce the need to generate costly real-world data, and provide additional training data to use in combination with lab-generated mass spectrometry data when training classifiers. Our approach is a novel combination of autoencoder-based synthetic data generation combined with a fixed, apriori defined hidden layer geometry. In particular, we train pairs of encoders and decoders with an additional loss term that enforces that the hidden layer passed from the encoder to the decoder match the embedding provided by an external deep learning model designed to predict functional properties of chemicals. We have verified that incorporating our synthetic spectra into a lab-generated dataset enhances the performance of classification algorithms compared to using only the real data. Our synthetic spectra have been successfully matched to lab-generated spectra for their respective chemicals using library matching software, further demonstrating the validity of our work. 

Abstract MF-LOGP, a new method for determining a single component octanol–water partition coefficients ($$LogP$$ $\mathrm{LogP}$) is presented which uses molecular formula as the only input. Octanol–water partition coefficients are useful in many applications, ranging from environmental fate and drug delivery. Currently, partition coefficients are either experimentally measured or predicted as a function of structural fragments, topological descriptors, or thermodynamic properties known or calculated from precise molecular structures. The MF-LOGP method presented here differs from classical methods as it does not require any structural information and uses molecular formula as the sole model input. MF-LOGP is therefore useful for situations in which the structure is unknown or where the use of a low dimensional, easily automatable, and computationally inexpensive calculations is required. MF-LOGP is a random forest algorithm that is trained and tested on 15,377 data points, using 10 features derived from the molecular formula to make$$LogP$$ $\mathrm{LogP}$predictions. Using an independent validation set of 2713 data points, MF-LOGP was found to have an average$$RMSE$$ $\mathrm{RMSE}$= 0.77 ± 0.007,$$MAE$$ $\mathrm{MAE}$= 0.52 ± 0.003, and$${R}^{2}$$ $R^2$= 0.83 ± 0.003. This performance fell within the spectrum of performances reported in the published literature for conventional higher dimensional models ($$RMSE$$ $\mathrm{RMSE}$= 0.42–1.54,$$MAE$$ $\mathrm{MAE}$= 0.09–1.07, and$${R}^{2}$$ $R^2$= 0.32–0.95). Compared with existing models, MF-LOGP requires a maximum of ten features and no structural information, thereby providing a practical and yet predictive tool. The development of MF-LOGP provides the groundwork for development of more physical prediction models leveraging big data analytical methods or complex multicomponent mixtures. Graphical Abstract 

Regression ensembles consisting of a collection of base regression models are often used to improve the estimation/prediction performance of a single regression model. It has been shown that the individual accuracy of the base models and the ensemble diversity are the two key factors affecting the performance of an ensemble. In this paper, we derive a theory for regression ensembles that illustrates the subtle trade-off between individual accuracy and ensemble diversity from the perspective of statistical correlations. Then, inspired by our derived theory, we further propose a novel loss function and a training algorithm for deep learning regression ensembles. We then demonstrate the advantage of our training approach over standard regression ensemble methods including random forest and gradient boosting regressors with both benchmark regression problems and chemical sensor problems involving analysis of Raman spectroscopy. Our key contribution is that our loss function and training algorithm is able to manage diversity explicitly in an ensemble, rather than merely allowing diversity to occur by happenstance.