Artificial intelligence (AI) specialist Unith (UNT) has called a trading halt in order to tap investors for some fresh funding. The company entered the trading halt on Thursday morning citing the planned capital raise, though it has not yet revealed the details of the raise. At this stage, shares will resume trade by market open on Monday morning, by which stage Unith plans to have announced how much it seeks to raise and how it will go about raising the funds. The company is fresh off a major rebranding in late-2022 when it changed its name from Crowd Media to Unith to better reflect the ongoing development of its "clean, commercial and agile" conversational AI technology. The company's core Talking Head technology is designed to humanise chatbots and improve business-human interactions.

The Artificial Intelligence/Machine Learning Consortium to Advance Health Equity and Researcher Diversity (AIM-AHEAD program) was created to combat harmful biases in how artificial intelligence and machine learning is used. KERA's Justin Martin talked with UNTHSC's Dr. Jamboor Vishwanatha, about what this means for North Texas. AIM-AHEAD is a consortium to promote artificial intelligence and machine learning to achieve health equity and also diversify the research workforce that is involved in the AI (artificial intelligence) and ML (machine learning) work. So it basically attacks two different issues. One is the lack of diversity in the data that is currently used in the AI/ML field.

DENTON, Texas (CBSDFW.COM) – The University of North Texas is launching a new program to meet the booming demand for artificial intelligence professionals. UNT will be soon offer the only Master of Science degree in AI in Texas. "I really think everyone should be learning more about AI," said Mark Albert, a computer science professor at UNT. From Google searches to chats with Alexa, AI is all around us, and it's expected to play a significant role in nearly every industry. "It's needed for a lot of applications, so we feel like we're in the position to help educate an AI-ready workforce, which is important for the state and the nation as well," said Yan Huang, senior associate dean of the UNT College of Engineering.

Artificial Intelligence applications are expanding into nearly every area of industry including government services, transportation, healthcare, cybersecurity, autonomous systems, finance and more. Forbes includes artificial intelligence as one of the "Hottest Career Paths of 2020 and Beyond." In order to meet the increasing demand for AI professionals, the University of North Texas, a Tier One research university, is offering the only Master of Science degree in artificial intelligence in Texas and one of only a few programs nationwide. The new degree offers students the choice of three concentrations: machine learning, autonomous systems and biomedical engineering. Students will be able to take classes that allow them to explore specific interests in AI and leave the program with marketable skills.

This paper proposes a learning strategy that embeds object-part concepts into a pre-trained convolutional neural network (CNN), in an attempt to 1) explore explicit semantics hidden in CNN units and 2) gradually transform the pre-trained CNN into a semantically interpretable graphical model for hierarchical object understanding. Given part annotations on very few (e.g., 3-12) objects, our method mines certain latent patterns from the pre-trained CNN and associates them with different semantic parts. We use a four-layer And-Or graph to organize the CNN units, so as to clarify their internal semantic hierarchy. Our method is guided by a small number of part annotations, and it achieves superior part-localization performance (about 13%-107% improvement in part center prediction on the PASCAL VOC and ImageNet datasets)

To ensure quality results from unreliable crowdsourced workers, task designers often construct complex workflows and aggregate worker responses from redundant runs. Frequently, they create several alternative workflows to accomplish the task, and choose a single workflow to deploy (perhaps the one that achieves the best performance during early experiments). However, this seemingly natural design paradigm does not achieve the full potential of crowdsourcing. In particular, using a single workflow (even the best) to accomplish a task is suboptimal. We show that alternative workflows can compose synergistically to yield a much higher quality output. We formalize the insight with a novel probabilistic graphical model, design and implement AgentHunt, a POMDP-based controller that dynamically switches between these workflows to achieve higher returns on investment, and design offline and online methods for learning model parameters. Live experiments on Amazon Mechanical Turk demonstrate the superiority of AgentHunt for the practical task of generating NLP training data, yielding up to 50% error reduction and greater net utility compared to previous methods.

To ensure quality results from unreliable crowdsourced workers, task designers often construct complex workflows and aggregate worker responses from redundant runs. Frequently, they experiment with several alternative workflows to accomplish the task, and eventually deploy the one that achieves the best performance during early trials. Surprisingly, this seemingly natural design paradigm does not achieve the full potential of crowdsourcing. In particular, using a single workflow (even the best) to accomplish a task is suboptimal. We show that alternative workflows can compose synergistically to yield much higher quality output. We formalize the insight with a novel probabilistic graphical model. Based on this model, we design and implement AGENTHUNT, a POMDP-based controller that dynamically switches between these workflows to achieve higher returns on investment. Additionally, we design offline and online methods for learning model parameters. Live experiments on Amazon Mechanical Turk demonstrate the superiority of AGENTHUNT for the task of generating NLP training data, yielding up to 50% error reduction and greater net utility compared to previous methods.