A Lagrangian Backward Air Parcel Trajectories Clustering Framework

Many studies concerning atmosphere moisture paths use Lagrangian backward air parcel trajectories to determine the humidity sources for specific locations. Automatically grouping trajectories according to their geographical position simplifies and speeds up their analysis. In this paper, we propose a framework for clustering Lagrangian backward air parcel trajectories, from trajectory generation to cluster accuracy evaluation. We employ a novel clustering algorithm, called DenLAC, to cluster troposphere air currents trajectories. Our main contribution is representing trajectories as a one-dimensional array consisting of each trajectory’s points position vector directions. We empirically test our pipeline by employing it on several Lagrangian backward trajectories initiated from Břeclav District, Czech Republic.

Genetic CFL: Hyperparameter Optimization in Clustered Federated Learning

Federated learning (FL) is a distributed model for deep learning that integrates client-server architecture, edge computing, and real-time intelligence. FL has the capability of revolutionizing machine learning (ML) but lacks in the practicality of implementation due to technological limitations, communication overhead, non-IID (independent and identically distributed) data, and privacy concerns. Training a ML model over heterogeneous non-IID data highly degrades the convergence rate and performance. The existing traditional and clustered FL algorithms exhibit two main limitations, including inefficient client training and static hyperparameter utilization. To overcome these limitations, we propose a novel hybrid algorithm, namely, genetic clustered FL (Genetic CFL), that clusters edge devices based on the training hyperparameters and genetically modifies the parameters clusterwise. Then, we introduce an algorithm that drastically increases the individual cluster accuracy by integrating the density-based clustering and genetic hyperparameter optimization. The results are bench-marked using MNIST handwritten digit dataset and the CIFAR-10 dataset. The proposed genetic CFL shows significant improvements and works well with realistic cases of non-IID and ambiguous data. An accuracy of 99.79% is observed in the MNIST dataset and 76.88% in CIFAR-10 dataset with only 10 training rounds.

High Accuracy Semi-Empirical Quantum Models Based on a Reduced Training Set

There exists a great need for computationally efficient quantum simulation approaches that can achieve an accuracy similar to high-level theories while exhibiting a wide degree of transferability. In this regard, we have leveraged a machine-learned force field based on Chebyshev polynomials to determine Density Functional Tight Binding (DFTB) models for organic materials. The benefit of our approach is two-fold: (1) many-body interactions can be corrected for in a systematic and rapidly tunable process, and (2) high-level quantum accuracy for a broad range of compounds can be achieved with ∼0.3% of data required for one advanced deep learning potential (ANI- 1x). In addition, the total number of data points in our training set is less than one half of that used for a recent DFTB-neural network model (trained on a separate dataset). Validation tests of our DFTB model against energy and vibrational data for gas-phase molecules for additional quantum datasets shows strong agreement with reference data from either hybrid density-functional theory, coupled-cluster calculations, or experiments. Preliminary testing on graphite and diamond successfully reproduce condensed phase structures. The models developed in this work, in principle, can retain most of the accuracy of quantum-based methods at any level of theory with relatively small training sets. Our efforts can thus allow for high throughput physical and chemical predictions with up to coupled-cluster accuracy for materials that are computationally intractable with standard approaches.

Indonesian Online News Extraction and Clustering Using Evolving Clustering

43,000 online media outlets in Indonesia publish at least one to two stories every hour. The amount of information exceeds human processing capacity, resulting in several impacts for humans, such as confusion and psychological pressure. This study proposes the Evolving Clustering method that continually adapts existing model knowledge in the real, ever-evolving environment without re-clustering the data. This study also proposes feature extraction with vector space-based stemming features to improve Indonesian language stemming. The application of the system consists of seven stages, (1) Data Acquisition, (2) Data Pipeline, (3) Keyword Feature Extraction, (4) Data Aggregation, (5) Predefined Cluster using Automatic Clustering algorithm, (6) Evolving Clustering, and (7) News Clustering Result. The experimental results show that Automatic Clustering generated 388 clusters as predefined clusters from 3.000 news. One of them is the unknown cluster. Evolving clustering runs for two days to cluster the news by streaming, resulting in a total of 611 clusters. Evolving clustering goes well, both updating models and adding models. The performance of the Evolving Clustering algorithm is quite good, as evidenced by the cluster accuracy value of 88%. However, some clusters are not right. It should be re-evaluated in the keyword feature extraction process to extract the appropriate features for grouping. In the future, this method can be developed further by adding other functions, updating and adding to the model, and evaluating.

Does Voicing Affect Patterns of Transfer in Nonnative Cluster Learning?

Purpose Previous studies have demonstrated that speakers can learn novel speech sequences, although the content and specificity of the learned speech motor representations remain incompletely understood. We investigated these representations by examining transfer of learning in the context of nonnative consonant clusters. Specifically, we investigated whether American English speakers who learn to produce either voiced or voiceless stop–stop clusters (e.g., /gd/ or /kt/) exhibit transfer to the other voicing pattern. Method Each participant ( n = 34) was trained on disyllabic nonwords beginning with either voiced (/gd/, /db/, /gb/) or voiceless (/kt/, /kp/, /tp/) onset consonant clusters (e.g., /gdimu/, /ktaksnæm/) in a practice-based speech motor learning paradigm. All participants were tested on both voiced and voiceless clusters at baseline (prior to practice) and in two retention sessions (20 min and 2 days after practice). We compared changes in cluster accuracy and burst-to-burst duration between baseline and each retention session to evaluate learning (performance on the trained clusters) and transfer (performance on the untrained clusters). Results Participants in both training conditions improved with respect to cluster accuracy and burst-to-burst duration for the clusters they practiced on. A bidirectional transfer pattern was found, such that participants also improved the cluster accuracy and burst-to-burst duration for the clusters with the other untrained voicing pattern. Post hoc analyses also revealed that improvement in the production of untrained stop–fricative clusters that originally were added as filler items. Conclusion Our findings suggest the learned speech motor representations may encode the information about the coordination of oral articulators for stop–stop clusters independently from information about the coordination of oral and laryngeal articulators.

Automatic identification of bird females using egg phenotype

Identification of individuals greatly contributes to understanding animal ecology and evolution, and in many cases can only be achieved using expensive and invasive techniques. Advances in computing technology offer alternative cost-effective techniques which are less invasive and can discriminate between individuals based on visual and/or acoustic cues. Here, we employ human assessment and an automatic analytical approach to predict the identity of common cuckoo (Cuculus canorus) females based on the appearance of their eggs. The cuckoo’s secretive brood parasitic strategy makes studying its life history very challenging. Eggs were analysed using calibrated digital photography for quantifying spotting pattern, size and shape, and spectrometry for measuring colour. Cuckoo females were identified from genetic sampling of their nestlings, allowing us to determine the accuracy of human and automatic female assignment. Finally, we used a novel ‘same-different’ approach that uses both genetic and phenotypic information to assign eggs that were not genetically analysed.Our results supported the ‘constant egg-type hypothesis’, showing that individual cuckoo females lay eggs with a relatively constant appearance and that eggs laid by different females differ more than eggs laid by the same female. The accuracy of unsupervised hierarchical clustering was comparable to assessments of experienced human observers. Supervised random forest analysis showed better results, with higher cluster accuracy. Same-different analysis was able to assign 22 of 87 unidentified cuckoo eggs to seven already known females.Our study showed that egg appearance on its own is not sufficient for identification of individual cuckoo females. We therefore advocate genetic analysis to be used for this purpose. However, supervised analytical methods reliably assigned a relatively high number of eggs without genetic data to their mothers which can be used in conjunction with genetic testing as a cost-effective method for increasing sample sizes for eggs where genetic samples could not be obtained.

Quantum chemical accuracy from density functional approximations via machine learning

Kohn-Sham density functional theory (DFT) is a standard tool in most branches of chemistry, but accuracies for many molecules are limited to 2-3 kcal ⋅ mol−1 with presently-available functionals. Ab initio methods, such as coupled-cluster, routinely produce much higher accuracy, but computational costs limit their application to small molecules. In this paper, we leverage machine learning to calculate coupled-cluster energies from DFT densities, reaching quantum chemical accuracy (errors below 1 kcal ⋅ mol−1) on test data. Moreover, density-based Δ-learning (learning only the correction to a standard DFT calculation, termed Δ-DFT ) significantly reduces the amount of training data required, particularly when molecular symmetries are included. The robustness of Δ-DFT is highlighted by correcting “on the fly” DFT-based molecular dynamics (MD) simulations of resorcinol (C6H4(OH)2) to obtain MD trajectories with coupled-cluster accuracy. We conclude, therefore, that Δ-DFT facilitates running gas-phase MD simulations with quantum chemical accuracy, even for strained geometries and conformer changes where standard DFT fails.