Oracle 1z0-1122-23 Oracle Cloud Infrastructure 2023 AI Foundations Associate Exam Practice Test

Tokens are the basic units of text representation in large language models. They can be words, subwords, characters, or symbols. Tokens are used to encode the input text into numerical vectors that can be processed by the model’s neural network. Tokens also determine the vocabulary size and the maximum sequence length of the model3. References: Oracle Cloud Infrastructure 2023 AI Foundations Associate | Oracle University

 Prompt engineering is the art of designing natural language instructions or queries that can elicit the desired response from a large language model. Prompt engineering does not modify the model parameters or architecture, but rather relies on the model’s existing knowledge and capabilities. Prompt engineering can be used to perform various tasks such as text generation, sentiment analysis, and code completion, by providing the model with the appropriate context, format, and constraints67. Prompt engineering is also known as zero-shot learning or query-based learning. References: [2211.01910] Large Language Models Are Human-Level Prompt Engineers] A developer’s guide to prompt engineering and LLMs - The GitHub Blog

Recurrent Neural Networks (RNNs) are a type of deep learning algorithm that can process sequential data, such as sentences, speech, or time series. They are composed of recurrent units that have a loop that allows them to store information from previous inputs and pass it to the next inputs. This way, they can capture the temporal dependencies and context within a sequence. RNNs can be used for various natural language processing tasks, such as text generation, machine translation, sentiment analysis, speech recognition, etc. However, RNNs also suffer from some limitations, such as vanishing or exploding gradients, difficulty in modeling long-term dependencies, and high computational cost. Therefore, some variants and extensions of RNNs have been proposed to overcome these challenges, such as Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), Bidirectional RNN (BiRNN), Attention Mechanism, etc. References: : [Recurrent neural network - Wikipedia], [What are Recurrent Neural Networks? | IBM], [Recurrent Neural Network (RNN) in Machine Learning]

Generative AI is a branch of artificial intelligence that focuses on creating new content or data based on the patterns and structure of existing data. Unlike other AI approaches that aim to recognize, classify, or predict data, generative AI aims to generate data that is realistic, diverse, and novel. Generative AI can produce various types of content, such as images, text, audio, video, software code, product designs, and more. Generative AI uses different techniques and models to learn from data and generate new examples, such as generative adversarial networks (GANs), variational autoencoders (VAEs), diffusion models, and foundation models. Generative AI has many applications across different domains and industries, such as art, entertainment, education, healthcare, engineering, marketing, and more. References: : Oracle Cloud Infrastructure AI - Generative AI, Generative artificial intelligence - Wikipedia

Large language models are trained on massive amounts of data to capture the complexity and diversity of natural language. Larger model sizes mean more parameters, which enable the model to learn more patterns and nuances from the data. Larger models also tend to generalize better to new tasks and domains. However, larger models also require more computational resources, data quality, and data size to train and deploy. Therefore, large language models handle the trade-off by prioritizing larger model sizes to achieve better performance, while using various techniques to optimize the training and inference efficiency4. References: Artificial Intelligence (AI) | Oracle

Fine-tuning is the process of updating the model parameters on a new task and dataset, using a pre-trained large language model as the starting point. Fine-tuning allows the model to adapt to the specific context and domain of the new task, and improve its performance and accuracy. Fine-tuning can be used to customize the model’s capabilities for specific tasks such as text classification, named entity recognition, and machine translation82. Fine-tuning is alsoknown as transfer learning or task-based learning. References: A Complete Guide to Fine Tuning Large Language Models, Finetuning Large Language Models - DeepLearning.AI

Multi-class classification is a type of supervised learning algorithm that is required in this scenario because the output variable has more than two classes. Multi-class classification is the problem of classifying instances into one of three or more classes. For example, classifying patients into low risk, moderate risk, or high risk based on their medical history and vital signs is a multi-class classification problem because each patient can only belong to one of these three classes. Multi-class classification can be solved by using various algorithms, such as decisiontrees, random forests, support vector machines (SVMs), k-nearest neighbors (k-NN), naive Bayes, logistic regression, neural networks, etc. Some of these algorithms can naturally handle multi-class problems, while others need to be adapted by using strategies such as one-vs-one or one-vs-rest. References: : Multiclass classification - Wikipedia, Multiclass Classification- Explained in Machine Learning

Oracle Cloud Infrastructure Language service is a cloud-based AI service for performing sophisticated text analysis at scale. It provides various capabilities to process unstructured text and extract structured information like sentiment or entities using natural language processing techniques. Some of the capabilities supported by Oracle Cloud Infrastructure Language service are:

• Language Detection: Detects languages based on the provided text, and includes a confidence score.
• Text Classification: Identifies the document category and subcategory that the text belongs to.
• Named Entity Recognition: Identifies common entities, people, places, locations, email, and so on.
• Key Phrase Extraction: Extracts an important set of phrases from a block of text.
• Sentiment Analysis: Identifies aspects from the provided text and classifies each into positive, negative, or neutral polarity.
• Text Translation: Translates text into the language of your choice.
• Personal Identifiable Information: Identifies, classifies, and de-identifies private information in unstructured text References: : Language Overview - Oracle, AI Text Analysis at Scale | Oracle

Reinforcement learning is a type of machine learning that is based on learning from outcomes to make decisions. Reinforcement learning algorithms learn from their own actions and experiences in an environment, rather than from labeled data or explicit feedback. The goal of reinforcement learning is to find an optimal policy that maximizes a cumulative reward over time. A policy is a rule that determines what action to take ineach state of the environment. A reward is a feedback signal that indicates how good or bad an action was for achieving a desired objective. Reinforcement learning involves a trial-and-error process of exploring different actions and observing their consequences, and then updating the policy accordingly. Some of the challenges and components of reinforcement learning are:

• Exploration vs exploitation: Balancing between trying new actions that might lead to higher rewards in the future (exploration) and choosing known actions that yield immediate rewards (exploitation).
• Markov decision process (MDP): A mathematical framework for modeling sequential decision making problems under uncertainty, where the outcomes depend only on the current state and action, not on the previous ones.
• Value function: A function that estimates the expected long-term return of each state or state-action pair, based on the current policy.
• Q-learning: A popular reinforcement learning algorithm that learns a value function called Q-function, which represents the quality of taking a certain action in a certain state.
• Deep reinforcement learning: A branch of reinforcement learning that combines deep neural networks with reinforcement learning algorithms to handle complex and high-dimensional problems, such as playing video games or controlling robots. References: : Reinforcement learning - Wikipedia, What is Reinforcement Learning? – Overview of How it Works - Synopsys