Browse Articles

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.

Plant diseases can cause up to 50% crop yield loss for the popular tomato plant. A mobile device-based method to identify diseases from photos of symptomatic leaves via computer vision can be more effective due to its convenience and accessibility. To enable a practical mobile solution, a “shallow” convolutional neural networks (CNNs) with few layers, and thus low computational requirement but with high accuracy similar to the deep CNNs is needed. In this work, we explored if such a model was possible.

Poor air quality is associated with negative effects on human health but can be difficult to measure in an accurate and cost-effective manner. The authors design and test a monitor for measuring indoor air quality using low-cost components.

In this study, the authors hypothesized that closed-circuit television images could be stored with improved resolution by using enhanced deep residual (EDSR) networks.

Improving the classifier of the difficulty level of Sudoku.

This study evaluates the potential of natural language processing (NLP) models in an emotion-driven bibliotherapy framework to improve mental health challenges.

Domestic waste classification using convolutional neural network

Based on the success of deep learning, recent works have attempted to develop a waste classification model using deep neural networks. This work presents federated learning (FL) for a solution, as it allows participants to aid in training the model using their own data. Results showed that with less clients, having a higher participation ratio resulted in less accuracy degradation by the data heterogeneity.

Machine translation remains a challenging area in artificial intelligence, with neural machine translation (NMT) making significant strides over the past decade but still facing hurdles, particularly in translation quality due to the reliance on expensive bilingual training data. This study explores whether large language models (LLMs), like GPT-4, can be effectively adapted for translation tasks and outperform traditional NMT systems.

Neural machine translation (NMT) is a software that uses neural network techniques to translate text from one language to another. However, one of the most famous NMT models—Google Translate—failed to give an accurate English translation of a famous Korean nursery rhyme, "Airplane" (비행기). The authors fine-tuned a pre-trained model first with a dataset from the lyrics domain, and then with a smaller dataset containing the rhythmical properties, to teach the model to translate rhythmically accurate lyrics. This stacked fine-tuning method resulted in an NMT model that could maintain the rhythmical characteristics of lyrics during translation while single fine-tuned models failed to do so.