Search By Subject

Installing and using machine learning software has become significantly more accessible over the years. The abundance of data, the affordability of computer power, and the availability of literature and quality data mining software have made it easier ...

Machine learning has shown its effectiveness in solving many problems for which traditional algorithmic solutions are not easy to find. For the past decade, we observed the rapid emergence of machine learning courses throughout the curricula. However, ...

This workshop will discuss the benefits and lessons learned in a project-oriented first course in software development. The intended audience for the workshop is similar to the makeup of the course under discussion - STEM-oriented students interested in ...

Participants will have an opportunity to learn the basic concepts of reinforcement learning. They can engage with a recently developed reinforcement learning system to play two different games, Connect 4 and Tic-Tac-Toe, with reinforcement learning ...

This paper focuses on the development of classifiers capable of detecting a skin cancer(s) given dermoscopic images. The dataset used for the training is a part of the 2019 ISIC Challenge, and consists of more than 25,000 labeled dermoscopic images. ...

Given a black-and-white photograph, recently proposed methods can generate a single plausible colorization. However, generated colorizations are often biased toward average colors, creating outputs that are muted and unnatural looking. Our proposed ...

Machine learning approaches learn models based on the statistical properties of training data. Learned models may be unfair due to bias inherent in the training data or because of spurious correlations based on sensitive attributes such as race or sex. ...

The utilization of neural networks has increased dramatically over the past decade, and yet our understanding of the effects of different network characteristics on efficacy is limited. Network architecture plays a significant role in their loss, but ...

Dr. Startiz will talk about some of the programs he worked on in the past, what makes robotics hard, and why the advances in AI are going to rapidly knock down those problems.

We document the linguistic structure of Russia's Internet Research Agency's social media and disinformation campaign to influence the 2016 U.S. Elections. Using the Discover Linguistic Inquiry and Word Count 2015 computerized text analysis tool, we ...

Among the many tools used for crime detection and prevention is that of machine learning techniques which shed light in the common quest of minimization and eradication of this negative action that affect the human society. We present a comparative ...

The history of computing and AI includes the largely forgotten Japanese Fifth Generation Computing Project in parallel knowledge-based AI that was announced at a sensitive time, capturing the imagination of the West (and the U.S.). After worldwide ...

Given the enormous cost of cybercrimes each year, many cybersecurity researchers are working to create an automated threat detection system, especially one that can accurately crawl non-English forums and markets. While text classification techniques ...

The field of Artificial Intelligence (AI) is certainly the buzz today, and we are deluged by TV and print ads from vendors offering AI hardware, software, and consulting services. This is rightly so as AI has achieved a high degree of success in recent ...

Breast cancer is the second most common cause of mortality in women in the United States [10]. An estimated 268,600 women were diagnosed with breast cancer in 2019, and 41,760 died from the disease [10]. The early detection of breast cancer is ...

In recent years, the number of available robotic platforms on the market has not only increased, but also the choices in terms of sensors, actuation, and language compatibility have diversified. While there is a large body of literature that discusses ...

When the computer science courses are designed, how to make the students understand the core concepts and algorithms is usually considered significantly. However, applying them to the real fields, which is considered more important, is often skipped. ...

Sentiment Analysis is a popular application of Natural Language Processing (NLP). This exercise offers the capability to perform opinion mining in the political arena by feeding data into a cloud natural language processor, without in-depth proficiency ...

Using artificial neural networks is an important approach for drawing inferences and making predictions when analyzing large and complex data sets. TensorFlow and PyTorch are two widely-used machine learning frameworks that support artificial neural ...

Tracking the state of a moving object is a challenging but useful application of computer vision. The problem is two-fold: efficiently and effectively locating the object within an image and deriving accurate state values despite measurement error or ...