Practice Exam: Oracle Cloud Infrastructure AI Foundations
1 In the context of machine learning, what does "overfitting" refer to?
Ans: A model that performs well on the training data but poorly on new data.
2.What is the primary goal of fine-tuning a Large Language Model (LLM)?
Ans: To adjust the pretrained model's parameters using a smaller, task-specific dataset, thereby improving its performance on specific tasks.
3. What is OCI Document Understanding primarily used for?
Ans: Automating document processing tasks
4 What is the role of the loss function in supervised learning algorithms?
Ans: It quantifies the cost of incorrect predictions.
5. Which of these is NOT a common application of unsupervised machine learning?
Ans: Spam detection
6. Which AI domain is primarily associated with the task of detecting and recognizing faces in images or videos?
Ans: Vision
7: What technique is used to predict the price of a house based on its features?
Ans: Regression
8. Which customization method can add latency to each request made to the LLM?
Ans: Retrieval Augmented Generation (RAG)
9. Which generative AI model has been particularly influential in Large Language Models (LLMs) for large-scale text generation and advanced natural language understanding?
Ans: Transformer-based models
10. Which optimization method requires a labeled dataset that can be most expensive and time-consuming to acquire compared to others?
Ans: Fine-tuning
11. When preparing data for a Machine Learning model, what are typically considered as input features?
Ans: Data collected after the model has made predictions
12. What does Retrieval Augmented Generation (RAG) involve?
Ans: Querying enterprise knowledge bases to provide grounded responses
13. Which type of model is most effective for sequence-to-sequence tasks such as machine translation?
Ans: Encoder-Decoder Transformer
14. Which algorithm is a nonparametric approach for supervised learning?
Ans: K-Nearest Neighbors (KNN)
15. What is one of the main challenges RNNs face when processing sequential data?
Ans: They struggle with understanding relationships between words that are far apart.
16. Which type of machine learning algorithm is used in predicting house prices?
Ans: Supervised Regression
17. In supervised learning, what does the term "hyperparameter" refer to?
Ans: A parameter set before training the model
18. What type of data is most likely to be used with Deep Learning algorithms?
Ans: Complex data with nonhuman interpretable features
19. How does Select AI enhance the interaction with Oracle Autonomous Database?
Ans: By enabling natural language prompts instead of SQL code
20. What is the main function of the encoder in a Transformer model?
Ans: To convert input text into embeddings
21. Which deep learning architecture is well-suited for processing sequential data such as sentences?
Ans: Recurrent Neural Network (RNN)
22. In the SQL Query Generation Process Flow, what is the second step after creating an AI Profile?
Ans: Specify Schemas and Tables
23. What is the primary goal of reinforcement learning?
Ans: To maximize cumulative reward
24. What is the purpose of the self-attention mechanism in Transformer-based models?
Ans: To model dependencies between different elements in a sequence
25. Which statement best describes the primary difference between Large Language Models (LLMs) and traditional Machine Learning (ML) models?
Ans: LLMs are pretrained on a large text corpus whereas ML models need to be trained on the custom data.
26. What is the primary capability of the OCI Language service?
Ans: Process unstructured data and extract insights
27. What role do tokens play in Large Language Models (LLMs)?
Ans: They are individual units into which a piece of text is divided during processing by the model.
28. What is the primary problem associated with vanishing gradients in Recurrent Neural Networks (RNNs)?
Ans: RNNs get extremely small gradients during backpropagation.
29. Which is NOT a part of Oracle Responsible AI guidelines?
Ans: Ensuring equality
30. What is the primary function of the inference process in machine learning?
Ans: Predicting outcomes from new data points
31. Question: world scenario?
Ans: Supervised learning is used for training models with labeled data, unsupervised learning is used for discovering hidden patterns in unlabeled data, and reinforcement learning is used for decision-making in dynamic environments.
32. Which task uses Generative AI?
Ans: Video generation
33. What is the role of a target variable in supervised learning?
Ans: It contains the desired output or class labels.
34. Which component of feedforward neural network is responsible for processing input data and forwarding it through various layers to make predictions?
Ans: Hidden Layer
35. An online bank wants to streamline the loan approval process. They have historical data on past loan applicants, including information on applicants' credit score, annual income, employment status, and whether they repaid the loan or defaulted.
Which machine learning algorithm can be used for this application?
Ans: Supervised Machine learning for classification
36. What is the primary role of Graphics Processing Units (GPUs) in modern computing?
Ans: Accelerating graphics rendering and parallel processing
37. You are writing poems. You need your computer to help you complete your lines by suggesting right words. Which Deep Learning model is well-suited for this task?
Ans: Recurrent Neural Network (RNN)
38. Which components are required to configure your data for natural language queries in Oracle Database 23ai?
Ans: Choose an LLM and Specify Schemas, Tables, and Views
39. Which AI domain can be used in the detection of fraudulent transactions?
Ans:
40. In the context of machine learning, what does "overfitting" refer to?
Ans: