Generative Ai Fundamentals Quiz Answers (3)

TQ generative AI MCQ: Test Your Knowledge!

 

 

What is 'Gradient Descent' in machine learning?
A technique for data visualization
An optimization algorithm for finding the minimum of a function
A method for increasing model complexity
A way to generate gradients in images
What is the main goal of image captioning in computer vision?
To classify images
To detect objects
To generate textual descriptions of images
To enhance image resolution
Which technology can help capture recent information not yet in foundation models?
Edge AI
Retrieval Augmented Generation
Multimodal Model
Large Action Model
Which US AI startup is known for its work on AI-powered protein folding prediction?
Deepmind (Alphabet subsidiary)
Anthropic
Recursion Pharmaceuticals
Insitro
What is the main function of DALL-E?
Text generation
Image generation from text descriptions
Speech recognition
Video editing
How can generative AI optimize business processes?
Summarizing content and outlining solution paths
Extracting and summarizing data for knowledge search functions
Generating synthetic data for supervised learning
Creating deepfakes for malicious purposes
How can you implement custom schedulers in PyTorch?
Only by using pre-defined schedulers
By subclassing _LRScheduler
By defining a new function
Both B and C
What is the key difference between a VAE and a standard autoencoder?
VAEs learn a probability distribution over the latent space
VAEs have a decoder component, while autoencoders do not
VAEs use convolutional layers, while autoencoders use fully connected layers
VAEs are generative models, while autoencoders are not
In the fashion industry, how might generative AI be utilized?
By replacing all fashion designers
By physically manufacturing all clothes
By managing all fashion shows
By assisting in design ideation and trend prediction
What challenges does generative AI face with respect to data?
No challenges at all
Potential copyright infringement issues
Only related to data storage
Exclusively about data encryption
What is the output of the generator in a GAN?
Realistic samples
Labels
Latent vectors
Loss function values
What is the purpose of a 'loss function' in machine learning?
To lose unnecessary data
To reduce processing power
To measure the model's performance
To create model limitations
What is the main contribution of SAGAN (Self-Attention GAN)?
To generate higher quality images
To incorporate self-attention mechanisms
To improve training speed
To reduce model size
Which of the following is a risk of AI without ethical considerations?
Increased accuracy
Fair treatment
Bias and discrimination
Improved efficiency
What is a method to improve AI transparency?
Hiding AI processes
Providing clear explanations of AI decision-making
Increasing AI autonomy
Minimizing human interaction
What is active learning in machine learning?
Learning while being active
Actively participating in learning
Learning about activities
A learning paradigm where the algorithm can query for labels
What is 'Stable Diffusion'?
A chemical process
A type of data compression
An open-source image generation AI
A method for stabilizing AI
LLMs can be used for language translation by:
Generating text word by word in the target language
Translating the text phrase by phrase
Encoding the input text and decoding it into the target language
Summarizing the text in the target language
What is adversarial machine learning?
Learning to be adversarial
Techniques to fool or manipulate AI systems
Adversarial roles in learning
Machine learning for adversaries
What is the primary purpose of a prompt engineering course?
To teach AI model architecture
To improve coding skills
To develop effective prompts for AI interactions
To design hardware for AI systems
How does GPT differ from BERT in terms of its architecture?
GPT uses bidirectional attention
BERT is decoder-only
GPT is encoder-only
GPT is unidirectional (left-to-right)
What are some challenges in using reinforcement learning for generative modeling?
Reward sparsity and credit assignment
Lack of reliable evaluation metrics
Computationally intensive training process
All of the above
Which of the following is a key challenge in using generative AI for dialogue systems?
Maintaining conversation coherence
Understanding context
Generating diverse responses
All of the above
How does the Longformer model differ from standard transformers in handling long documents?
It uses a combination of sliding window and global attention
It implements hierarchical attention
It uses compression techniques
It employs dynamic context selection
BARD, the conversational AI chatbot, is developed by which company?
Open AI
Google
Amazon
Microsoft
Score: 0/25