Scribd
Upload
2 views2 pages

Assignment

The assignment requires groups of five students to answer ten questions related to supervised learning, AI systems, and agentic AI, with a submission deadline of April 12, 2026, in PDF format. Topics include handling noisy data, feature selection for crop yield prediction, underfitting vs overfitting, and the implications of bias in training data. Students must provide practical examples and discuss the integration of AI models in various scenarios.

Uploaded by

Ebrahim Beker
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
2 views2 pages

Assignment

The assignment requires groups of five students to answer ten questions related to supervised learning, AI systems, and agentic AI, with a submission deadline of April 12, 2026, in PDF format. Topics include handling noisy data, feature selection for crop yield prediction, underfitting vs overfitting, and the implications of bias in training data. Students must provide practical examples and discuss the integration of AI models in various scenarios.

Uploaded by

Ebrahim Beker
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Selected Topics – Assignment (15 Marks)

Make group of 5 members and do the assignment

Submission Format: Students shall submit their answers in PDF


format.

Deadline: April 12,2026

Instructions: Answer all questions.

1. Explain how a supervised learning model can handle noisy data


in a training dataset. Provide a practical example relevant to
image or text data.
2. You are designing an AI system to predict crop yields. How
would you choose the features for a supervised learning model,
and why would feature selection matter?
3. Differentiate between underfitting and overfitting in supervised
learning. Suggest one technique to reduce each problem in a real-
world AI application.
4. Describe a scenario where agentic AI could autonomously make
decisions to optimize energy consumption in a smart building.
What type of feedback or learning mechanism would be
necessary for the AI to improve over time?
5. Explain the difference between supervised learning and
reinforcement learning. Why is agentic AI more aligned with
reinforcement learning principles?
6. Imagine you have a dataset with labeled customer reviews. How
would a supervised learning model help in sentiment analysis,
and what challenges could arise if the dataset contains ambiguous
labels?
7. Define agentic AI in your own words. Provide one example where
it could assist humans in daily decision-making without constant
supervision.
8. Consider an AI model that predicts student performance. Discuss
how bias in the training data could affect outcomes, and propose
a method to mitigate it.
9. In supervised learning, loss functions guide model training.
Explain why choosing an appropriate loss function is critical and
give an example where a common loss function might fail.
10. Imagine a hybrid system where agentic AI is integrated with
a supervised learning model for medical diagnosis. Discuss one
potential benefit and one potential risk of this combination.

You might also like