Browse Articles

Hybrid Quantum-Classical Generative Adversarial Network for synthesizing chemically feasible molecules

Sikdar et al. | Jan 10, 2023

Hybrid Quantum-Classical Generative Adversarial Network for synthesizing chemically feasible molecules

Current drug discovery processes can cost billions of dollars and usually take five to ten years. People have been researching and implementing various computational approaches to search for molecules and compounds from the chemical space, which can be on the order of 1060 molecules. One solution involves deep generative models, which are artificial intelligence models that learn from nonlinear data by modeling the probability distribution of chemical structures and creating similar data points from the trends it identifies. Aiming for faster runtime and greater robustness when analyzing high-dimensional data, we designed and implemented a Hybrid Quantum-Classical Generative Adversarial Network (QGAN) to synthesize molecules.

Read More...

Prediction of molecular energy using Coulomb matrix and Graph Neural Network

Hazra et al. | Feb 01, 2022

Prediction of molecular energy using Coulomb matrix and Graph Neural Network

With molecular energy being an integral element to the study of molecules and molecular interactions, computational methods to determine molecular energy are used for the preservation of time and resources. However, these computational methods have high demand for computer resources, limiting their widespread feasibility. The authors of this study employed machine learning to address this disadvantage, utilizing neural networks trained on different representations of molecules to predict molecular properties without the requirement of computationally-intensive processing. In their findings, the authors determined the Feedforward Neural Network, trained by two separate models, as capable of predicting molecular energy with limited prediction error.

Read More...

The Effect of Varying Training on Neural Network Weights and Visualizations

Fountain et al. | Dec 04, 2019

The Effect of Varying Training on Neural Network Weights and Visualizations

Neural networks are used throughout modern society to solve many problems commonly thought of as impossible for computers. Fountain and Rasmus designed a convolutional neural network and ran it with varying levels of training to see if consistent, accurate, and precise changes or patterns could be observed. They found that training introduced and strengthened patterns in the weights and visualizations, the patterns observed may not be consistent between all neural networks.

Read More...

Open Source RNN designed for text generation is capable of composing music similar to Baroque composers

Goel et al. | May 05, 2021

Open Source RNN designed for text generation is capable of composing music similar to Baroque composers

Recurrent neural networks (RNNs) are useful for text generation since they can generate outputs in the context of previous ones. Baroque music and language are similar, as every word or note exists in context with others, and they both follow strict rules. The authors hypothesized that if we represent music in a text format, an RNN designed to generate language could train on it and create music structurally similar to Bach’s. They found that the music generated by our RNN shared a similar structure with Bach’s music in the input dataset, while Bachbot’s outputs are significantly different from this experiment’s outputs and thus are less similar to Bach’s repertoire compared to our algorithm.

Read More...

Identifying Neural Networks that Implement a Simple Spatial Concept

Zirvi et al. | Sep 13, 2022

Identifying Neural Networks that Implement a Simple Spatial Concept

Modern artificial neural networks have been remarkably successful in various applications, from speech recognition to computer vision. However, it remains less clear whether they can implement abstract concepts, which are essential to generalization and understanding. To address this problem, the authors investigated the above vs. below task, a simple concept-based task that honeybees can solve, using a conventional neural network. They found that networks achieved 100% test accuracy when a visual target was presented below a black bar, however only 50% test accuracy when a visual target was presented below a reference shape.

Read More...

The Effect of Various Liquid Mediums on the Transport of Photonic Energy and its Impact on the Quantum Efficiency of Photovoltaic Cells

Payra et al. | May 05, 2015

The Effect of Various Liquid Mediums on the Transport of Photonic Energy and its Impact on the Quantum Efficiency of Photovoltaic Cells

A photovoltaic cell (PV cell), or solar cell, converts the energy of light into electricity and is the basis for solar power. In order to increase the efficiency of PV cells, the authors in this study used common household items as photon transmissions mediums and measured their effects on the temperature and voltage output of the PV cells.

Read More...

Artificial Intelligence Networks Towards Learning Without Forgetting

Kreiman et al. | Oct 26, 2018

Artificial Intelligence Networks Towards Learning Without Forgetting

In their paper, Kreiman et al. examined what it takes for an artificial neural network to be able to perform well on a new task without forgetting its previous knowledge. By comparing methods that stop task forgetting, they found that longer training times and maintenance of the most important connections in a particular task while training on a new one helped the neural network maintain its performance on both tasks. The authors hope that this proof-of-principle research will someday contribute to artificial intelligence that better mimics natural human intelligence.

Read More...

Computational development of aryl sulfone compounds as potential NNRTIs

Zhang et al. | Oct 12, 2022

Computational development of aryl sulfone compounds as potential NNRTIs

Non-nucleoside reverse transcriptase inhibitors (NNRTIs) are allosteric inhibitors that bind to the HIV reverse transcriptase and prevent replication. Indolyl aryl sulfones (IAS) and IAS derivatives have been found to be highly effective against mutant strains of HIV-1 reverse transcriptase. Here, we analyzed molecules designed using aryl sulfone scaffolds paired to cyclic compounds as potential NNRTIs through the computational design and docking of 100 novel NNRTI candidates. Moreover, we explored the specific combinations of functional groups and aryl sulfones that resulted in the NNRTI candidates with the strongest binding affinity while testing all compounds for carcinogenicity. We hypothesized that the combination of an IAS scaffold and pyrimidine would produce the compounds with the best binding affinity. Our hypothesis was correct as the series of molecules with an IAS scaffold and pyrimidine exhibited the best average binding affinity. Additionally, this study found 32 molecules designed in this procedure with higher or equal binding affinities to the previously successful IAS derivative 5-bromo-3-[(3,5-dimethylphenyl)sulfonyl]indole-2-carboxyamide when docked to HIV-1 reverse transcriptase.

Read More...

Search Articles

Search articles by title, author name, or tags

Clear all filters

Popular Tags

Browse by school level