Human gesture classification by brute-force machine learning for exergaming in physiotherapy

<div>MOFs and COFs are porous materials with a large variety of applications including gas</div><div>storage and separation. Synthesised in a modular fashion from distinct building blocks, a</div><div>near in?nite number of structures can be constructed and the properties of the material can</div><div>be tailored for a speci?c application. While this modularity is a very attractive feature it also</div><div>poses a challenge. Attempting to identify the best performing material(s) for a given appli-</div><div>cation is experimentally intractable. Current research e?orts combine molecular simulations</div><div>and machine learning techniques to evaluate the simulated performance of hundreds of thou-</div><div>sands of materials to identify top performing MOFs and COFs for a given application. These</div><div>approaches typically rely on moderated brute-force screening which is still resource-intensive</div><div>as typically between 70 - 100 % of the hundreds of thousands of materials must be simulated</div><div>to create a training set for the machine learning models used, restricting screening to rela-</div><div>tively simple molecules. In this work we demonstrate our novel Bayesian mining approach</div><div>to materials screening which allows 62 - 92 % of the top 100 porous materials for a range of</div><div>applications to be readily identi?ed from large materials databases after only assessing less</div><div>than one percent of all materials. This is a stark contrast to the 0 - 1 % achieved by conven-</div><div>tional brute-force screening where porous materials are just chosen at random during a high</div><div>throughput screening. Through this accelerated virtual screening process, the identi?cation of</div><div>high performing materials can be used to more rapidly inform experimental e?orts and hence</div><div>lead to an acceleration of the entire research and development pipeline of porous materials.</div>

Download Full-text

Gesture Classification Using Machine Learning Approach

International Journal of Mechanical and Production Engineering Research and Development ◽

10.24247/ijmperdjun20201280 ◽

2020 ◽

Vol 10 (3) ◽

pp. 13435-13440

Author(s):

Ketki P. Kshirsagar et al., Ketki P. Kshirsagar et al., ◽

Keyword(s):

Machine Learning ◽

Learning Approach ◽

Machine Learning Approach ◽

Gesture Classification

Download Full-text

Build Orientation Optimization for Strength Enhancement of FDM Parts Using Machine Learning based Algorithm

10.31224/osf.io/3dh9s ◽

2019 ◽

Author(s):

Manoj Malviya ◽

Kaushal A Desai

Keyword(s):

Machine Learning ◽

Brute Force ◽

Initial Sample ◽

Bayesian Algorithm ◽

Build Orientation ◽

Optimization Framework ◽

Sample Data ◽

Artificial Neural Network Ann ◽

Brute Force Approach ◽

Orientation Optimization

The layered fabrication approach induces directional anisotropy and impacts mechanical strength of FDM components significantly. This paper proposes generalized machine learning based parameter optimization framework to determine optimal build orientation for FDM components. The algorithm determines ideal build orientation by maximizing the minimum Factor of Safety (FoS) for the component under prescribed loading conditions ensuring its even distribution. An Artificial Neural Network (ANN) coupled with Bayesian algorithm has been employed to accelerate the optimization process. The algorithm begins with an initial sample data collected using brute force approach; uses single layered ANN for approximation and optimization is achieved using Bayesian algorithm. A series of computational experiments considering five different test components has been devised to evaluate the performance and efficacy of the proposed algorithm. These experiments demonstrated that the proposed algorithm can determine the optimum building orientation effectively with certain limitations

Download Full-text

Feature partitioning for robust tree ensembles and their certification in adversarial scenarios

EURASIP Journal on Information Security ◽

10.1186/s13635-021-00127-0 ◽

2021 ◽

Vol 2021 (1) ◽

Author(s):

Stefano Calzavara ◽

Claudio Lucchese ◽

Federico Marcuzzi ◽

Salvatore Orlando

Keyword(s):

Machine Learning ◽

Machine Learning Algorithms ◽

Brute Force ◽

Accuracy Improvement ◽

Safe Environment ◽

Feature Partitioning ◽

Feature Based ◽

Brute Force Approach ◽

The Given ◽

Certification Method

AbstractMachine learning algorithms, however effective, are known to be vulnerable in adversarial scenarios where a malicious user may inject manipulated instances. In this work, we focus on evasion attacks, where a model is trained in a safe environment and exposed to attacks at inference time. The attacker aims at finding a perturbation of an instance that changes the model outcome.We propose a model-agnostic strategy that builds a robust ensemble by training its basic models on feature-based partitions of the given dataset. Our algorithm guarantees that the majority of the models in the ensemble cannot be affected by the attacker. We apply the proposed strategy to decision tree ensembles, and we also propose an approximate certification method for tree ensembles that efficiently provides a lower bound of the accuracy of a forest in the presence of attacks on a given dataset avoiding the costly computation of evasion attacks.Experimental evaluation on publicly available datasets shows that the proposed feature partitioning strategy provides a significant accuracy improvement with respect to competitor algorithms and that the proposed certification method allows ones to accurately estimate the effectiveness of a classifier where the brute-force approach would be unfeasible.

Download Full-text

High-throughput Calculation of Atomic Planar Density for Compounds

10.33774/chemrxiv-2021-l9rp7 ◽

2021 ◽

Author(s):

Sterling Baird ◽

Taylor Sparks

Keyword(s):

Machine Learning ◽

Computational Geometry ◽

Open Source ◽

High Throughput ◽

Lattice Plane ◽

Brute Force ◽

Large Collection

A large collection of element-wise planar densities for compounds obtained from the Materials Project is calculated using brute force computational geometry methods. We demonstrate that the element-wise max lattice plane densities can be useful as machine learning features. The methods described here are implemented in an open-source Mathematica package hosted at https://github.com/sgbaird/LatticePlane.

Download Full-text