Batch Mode Active Learning for Regression With Expected Model Change

ANI-1ccx-gelu Universal Interatomic Potential and Its Fine-Tuning: Toward Accurate and Efficient Anharmonic Vibrational Frequencies

Calculating anharmonic vibrational modes of molecules for interpreting experimental spectra is one of the most interesting challenges of contemporary computational chemistry. However, the traditional QM methods are costly for this application. Machine learning techniques have emerged as a powerful tool for substituting the traditional QM methods. Universal interatomic potentials (UIPs) hold a particular promise to deliver accurate results at a fraction of the cost of the traditional QM methods, but the performance of UIPs for calculating anharmonic vibrational frequencies remains hitherto unknown. Here we show that despite a known excellent performance of the representative UIP ANI-1ccx for thermochemical properties, it fails for the anharmonic frequencies due to the original unfortunate choice of the activation function. Hence, we recommend evaluating new UIPs on anharmonic frequencies as an additional important quality test. To remedy the shortcomings of ANI-1ccx, we introduce its reformulation ANI-1ccx-gelu with the GELU activation function, which is capable of calculating IR anharmonic frequencies with reasonable accuracy (close to B3LYP/6-31G*). We also show that our new UIP can be fine-tuned to obtain very accurate anharmonic frequencies for some specific molecules but more effort is needed to improve the overall quality of UIP and its capability for fine-tuning. The new UIP will be included as part of our universal and updatable AI-enhanced QM methods (UAIQM) platform and is available together with usage and fine-tuning tutorials in open-source MLatom at https://github.com/dralgroup/mlatom. The calculations can also be performed via a web browser at https://XACScloud.com.

AutoML based workflow for design of experiments (DOE) selection and benchmarking data acquisition strategies with simulation models

Design of experiments (DOE) is an established method to allocate resources for efficient parameter space exploration. Model based active learning (AL) data sampling strategies have shown potential for further optimization. This paper introduces a workflow for conducting DOE comparative studies using automated machine learning. Based on a practical definition of model complexity in the context of machine learning, the interplay of systematic data generation and model performance is examined considering various sources of uncertainty: this includes uncertainties caused by stochastic sampling strategies, imprecise data, suboptimal modeling, and model evaluation. Results obtained from electrical circuit models with varying complexity show that not all AL sampling strategies outperform conventional DOE strategies, depending on the available data volume, the complexity of the dataset, and data uncertainties. Trade-offs in resource allocation strategies, in particular between identical replication of data points for statistical noise reduction and broad sampling for maximum parameter space exploration, and their impact on subsequent machine learning analysis are systematically investigated. Results indicate that replication oriented strategies should not be dismissed but may prove advantageous for cases with non-negligible noise impact and intermediate resource availability. The provided workflow can be used to simulate practical experimental conditions for DOE testing and DOE selection.

Active Learning Query Strategies for Linear Regression Based on Efficient Global Optimization

Active learning, a subfield of machine learning, can train a good model by selecting a minimum number of labeled samples. In many machine learning scenarios, needed information (such as the best value in unlabeled datasets) is acquired by prediction. When there is too little data in the training model, the prediction accuracy would obviously affect the accuracy of the results. To establish a high-performance regression model for a small dataset while accelerating the search for the best sample, a new active learning query strategy, EGO-ALR, that combines efficient global optimization (EGO) and active learning for regression (ALR) was proposed. It was found that the performance of EGO-ALR was significantly better than the original ALR query strategies in terms of the root mean square error (RMSE), correlation coefficient (CC), and opportunity cost (Oppo Cost). Specifically, EGO-ALR increased the Oppo Cost by an average of 25.27% when the RMSE and CC values were not more than 1.07% different from the original ALR. This study validated the efficiency and robustness of EGO-ALR approaches using 19 datasets from various domains and three distinct linear regression models (Ridge regression, Lasso, and Elastic network).

An Online Security Prediction and Control Framework for Modern Power Grids

The proliferation of renewable energy sources distributed generation (RES-DG) into the grid results in time-varying inertia constant. To ensure the security of the grid under varying inertia, techniques for fast security assessment are required. In addition, considering the high penetration of RES-DG units into the modern grids, security prediction using varying grid features is crucial. The computation burden concerns of conventional time-domain security assessment techniques make it unsuitable for real-time security prediction. This paper, therefore, proposes a fast security monitoring model that includes security prediction and load shedding for security control. The attributes considered in this paper include the load level, inertia constant, fault location, and power dispatched from the renewable energy sources generator. An incremental Naïve Bayes algorithm is applied on the training dataset developed from the responses of the grid to transient stability simulations. An additive Gaussian process regression (GPR) model is proposed to estimate the load shedding required for the predicted insecure states. Finally, an algorithm based on the nodes’ security margin is proposed to determine the optimal node (s) for the load shedding. The average security prediction and load shedding estimation model training times are 1.2 s and 3 s, respectively. The result shows that the proposed model can predict the security of the grid, estimate the amount of load shed required, and determine the specific node for load shedding operation.

Embedding Perspective Analysis Into Multi-Column Convolutional Neural Network for Crowd Counting

The crowd counting is challenging for deep networks due to several factors. For instance, the networks can not efficiently analyze the perspective information of arbitrary scenes, and they are naturally inefficient to handle the scale variations. In this work, we deliver a simple yet efficient multi-column network, which integrates the perspective analysis method with the counting network. The proposed method explicitly excavates the perspective information and drives the counting network to analyze the scenes. More concretely, we explore the perspective information from the estimated density maps and quantify the perspective space into several separate scenes. We then embed the perspective analysis into the multi-column framework with a recurrent connection. Therefore, the proposed network matches various scales with the different receptive fields efficiently. Secondly, we share the parameters of the branches with various receptive fields. This strategy drives the convolutional kernels to be sensitive to the instances with various scales. Furthermore, to improve the evaluation accuracy of the column with a large receptive field, we propose a transform dilated convolution. The transform dilated convolution breaks the fixed sampling structure of the deep network. Moreover, it needs no extra parameters and training, and the offsets are constrained in a local region, which is designed for the congested scenes. The proposed method achieves state-of-the-art performance on five datasets (ShanghaiTech, UCF CC 50, WorldEXPO'10, UCSD, and TRANCOS).

Pool-Based Sequential Active Learning for Regression

Active learning (AL) is a machine-learning approach for reducing the data labeling effort. Given a pool of unlabeled samples, it tries to select the most useful ones to label so that a model built from them can achieve the best possible performance. This paper focuses on pool-based sequential AL for regression (ALR). We first propose three essential criteria that an ALR approach should consider in selecting the most useful unlabeled samples: informativeness, representativeness, and diversity, and compare four existing ALR approaches against them. We then propose a new ALR approach using passive sampling, which considers both the representativeness and the diversity in both the initialization and subsequent iterations. Remarkably, this approach can also be integrated with other existing ALR approaches in the literature to further improve the performance. Extensive experiments on 11 University of California, Irvine, Carnegie Mellon University StatLib, and University of Florida Media Core data sets from various domains verified the effectiveness of our proposed ALR approaches.

Stopping Criterion for Active Learning with Model Stability

Active learning selectively labels the most informative instances, aiming to reduce the cost of data annotation. While much effort has been devoted to active sampling functions, relatively limited attention has been paid to when the learning process should stop. In this article, we focus on the stopping criterion of active learning and propose a model stability--based criterion, that is, when a model does not change with inclusion of additional training instances. The challenge lies in how to measure the model change without labeling additional instances and training new models. Inspired by the stochastic gradient update rule, we use the gradient of the loss function at each candidate example to measure its effect on model change. We propose to stop active learning when the model change brought by any of the remaining unlabeled examples is lower than a given threshold. We apply the proposed stopping criterion to two popular classifiers: logistic regression (LR) and support vector machines (SVMs). In addition, we theoretically analyze the stability and generalization ability of the model obtained by our stopping criterion. Substantial experiments on various UCI benchmark datasets and ImageNet datasets have demonstrated that the proposed approach is highly effective.