Which fiber optic connector type is known for its round shape and “twist-to-lock” (bayonet-style) mechanism?

A technician is running network cables through the space above a drop ceiling that is used for air circulation (HVAC). To comply with fire safety codes, which type of cable jacket must be used?

EMERGING TECHNOLOGY TOPIC Prior to 1926, people had to manually open their garage doors.  In 1926, C.G. Johnson invented the first powered garage door opener.  What kind of innovation was the powered garage door opener when it came to the market in 1926?

EMERGING TECHNOLOGY TOPIC Coca-Cola recently released Coca-Cola Orange Cream in February 2025. This new flavor was designed by analyzing consumer trends and Freestyle machine choices across different markets, showing the brand’s ongoing experimentation with flavored variants for nostalgic tastes. What kind of innovation was the release of Coca-Cola Orange Cream beverage?

BUSINESS INTELLIGENCE TOPIC What type of knowledge is defined as being gained through experience and is intuitively understood?  

EITHICS TOPIC From the Ethics topic additional terms and definitions slides, what is the ethical concern today related to Section 230 – Title 47 of U.S. Code Telecommunications Act? 

ARTIFICAL INTELLIGENCE TOPIC For each statement below select the best augmented generation method (RAG or CAG) which enhances the capabilities of large language models(LLMs). 1- Retrieve only the things we think we will need: [One] 2- Retrieve everything upfront and remember it for later: [Two] 3- Create a model with high accuracy to specific data: [Three] 4- Create a model with very low latency: [Four] 5- Create a model which can change easily to be fresh: [Five] 6- Create a model which is highly scalable with unlimited datasets: [Six]

DIGITAL TRANSFORMATION TOPIC Which of the following are part of the Bain & Company’s Digital Transformation framework? (multiple-select)

ETHICS TOPIC When making ethical decisions what is the most preferred decision workflow order? 

Question 8: (12 points) Consider the best practices for training neural networks. Answer the following questions: (3 points) In practice, the Nesterov momentum often converges faster than the standard momentum. Explain why the correction based on the gradient at the anticipated position might prevent overshooting. (6 points) You observe the following training behaviors: Observation 1: Your deep network (8 layers) trains very slowly. The gradients in early layers are extremely small. Training loss decreases but very gradually. Observation 2: Your network achieves 99\% training accuracy but only 70\% validation accuracy. The gap is large and consistent. Observation 3: Your network’s training is unstable – loss fluctuates wildly and sometimes diverges. Different random initializations lead to very different outcomes. For each observation: (i) Identify whether dropout, batch normalization, or both would help, and (ii) Explain why the chosen technique addresses the specific problem.   3. (3 points) Consider a feedforward neural network with the following architecture: Input (100 features) → Dense(256) → ReLU → Dense(128) → ReLU → Dense(10) → Softmax. Calculate the total number of trainable parameters in this network. Show your work by computing the parameters for each layer separately, including both weights and biases.