1Z0-1122
Coming Soon
Expected availability announced soon
This course is in active development. Preview the scope below and create a free account to be notified the moment it goes live.
OCI AI Foundations
The Oracle Cloud Infrastructure AI Foundations Associate exam validates understanding of core AI/ML concepts, deep learning, generative AI, and OCI AI services, enabling professionals to design and deploy AI solutions on OCI.
90
Minutes
50
Questions
68
Passing Score
Who Should Take This
It is ideal for cloud engineers, data scientists, and solution architects who have foundational experience with cloud platforms and a basic grasp of machine learning. These professionals seek to demonstrate competency in OCI AI services and generative AI to advance their careers and support AI‑driven initiatives within their organizations.
What's Covered
1
AI and Machine Learning Concepts
2
Deep Learning and Neural Networks
3
Generative AI Concepts
4
OCI AI and ML Services
5
OCI Generative AI Service
What's Included in AccelaStudy® AI
Adaptive Knowledge Graph
Practice Questions
Lesson Modules
Console Simulator Labs
Exam Tips & Strategy
20 Activity Formats
Course Outline
60 learning goals
1
AI and Machine Learning Concepts
2 topics
AI Fundamentals
- Define artificial intelligence, machine learning, and deep learning and explain the hierarchical relationship among these fields.
- Identify common AI application categories including recommendation systems, fraud detection, image recognition, NLP, and predictive analytics.
- Explain responsible AI principles including fairness, transparency, privacy, robustness, and accountability for ethical AI development.
- Describe the differences between narrow AI and general AI and identify current AI capability boundaries in enterprise applications.
- Explain the role of data quality, quantity, and diversity in determining AI system accuracy and reliability.
Machine Learning Fundamentals
- Differentiate between supervised, unsupervised, semi-supervised, and reinforcement learning paradigms and their data requirements.
- Identify common classification algorithms: logistic regression, decision trees, random forests, support vector machines, and naive Bayes.
- Identify common regression and clustering algorithms: linear regression, polynomial regression, k-means, hierarchical clustering, and DBSCAN.
- Explain the ML workflow from data collection through feature engineering, model training, validation, evaluation, and deployment.
- Interpret model evaluation metrics: accuracy, precision, recall, F1 score, AUC-ROC for classification; MSE and RMSE for regression.
- Describe data challenges including class imbalance, missing values, overfitting, underfitting, and feature selection impacts on model quality.
- Analyze a business scenario to determine the most appropriate ML paradigm and algorithm category based on data and desired outcomes.
2
Deep Learning and Neural Networks
3 topics
Neural Network Architecture
- Describe neural network components: neurons, layers (input, hidden, output), weights, biases, and common activation functions (ReLU, sigmoid, softmax).
- Explain forward propagation, loss computation, backpropagation, and gradient descent variants (SGD, Adam) for training neural networks.
- Identify neural network types: feedforward, convolutional (CNN), recurrent (RNN), long short-term memory (LSTM), and transformers.
- Describe training concepts including epochs, batch size, learning rate scheduling, early stopping, and regularization (dropout, L2).
Computer Vision
- Identify computer vision tasks: image classification, object detection, semantic segmentation, instance segmentation, and OCR.
- Describe CNN components including convolutional layers, pooling layers, and fully connected layers for hierarchical feature extraction.
- Explain transfer learning and pre-trained models (ResNet, VGG, YOLO) for building CV applications with limited training data.
- Analyze a visual data processing scenario to select the appropriate CV technique based on task requirements and data characteristics.
Natural Language Processing
- Identify NLP tasks: text classification, sentiment analysis, NER, machine translation, summarization, and question answering.
- Explain tokenization methods (word, subword, BPE) and their impact on vocabulary size and out-of-vocabulary handling.
- Describe word embeddings (Word2Vec, GloVe) and contextual embeddings (BERT, GPT) for numerical text representation.
- Compare bag-of-words, TF-IDF, and embedding-based approaches for text representation and identify when each is most appropriate.
3
Generative AI Concepts
3 topics
Large Language Models and Transformers
- Describe transformer architecture: self-attention, multi-head attention, encoder-decoder structure, positional encoding, and layer normalization.
- Explain LLM pre-training on large corpora followed by fine-tuning and RLHF alignment for producing helpful, harmless responses.
- Describe prompt engineering: zero-shot, few-shot, chain-of-thought, system prompts, temperature, top-p, and frequency penalties.
- Explain retrieval augmented generation (RAG): vector stores, embedding models, similarity search, context injection, and grounding.
- Describe fine-tuning approaches: full fine-tuning, LoRA, QLoRA, and prompt tuning for adapting foundation models to domains.
Generative AI Applications and Risks
- Identify generative AI modalities: text generation, code generation, image generation (diffusion models), audio synthesis, and video generation.
- Explain generative AI risks: hallucinations, bias amplification, toxicity, data leakage, copyright concerns, and prompt injection attacks.
- Describe evaluation methods: human evaluation, BLEU, ROUGE, perplexity, and benchmark suites for assessing generative model quality.
- Analyze a generative AI use case to evaluate feasibility, identify risks, and recommend guardrails for responsible deployment.
Multimodal and Emerging AI
- Describe multimodal AI models that process text, images, and audio simultaneously for tasks like visual question answering.
- Explain AI agent architectures including tool use, planning, memory, and chain-of-action patterns for autonomous task completion.
- Identify emerging AI paradigms including federated learning, self-supervised learning, and neural architecture search and their potential applications.
- Analyze a scenario requiring multiple AI modalities to recommend an integrated approach combining vision, language, and structured data services.
4
OCI AI and ML Services
2 topics
OCI Pre-built AI Services
- Describe OCI Vision service for image classification, object detection, and document AI for automated visual data processing.
- Describe OCI Speech service for automatic speech recognition and real-time transcription capabilities.
- Explain OCI Language service for sentiment analysis, entity extraction, key phrase extraction, language detection, and text classification.
- Explain OCI Document Understanding for extracting tables, key-value pairs, and text from invoices, receipts, and structured forms.
- Describe OCI Anomaly Detection for identifying unusual patterns in multivariate time-series data for fraud and equipment monitoring.
- Analyze a business problem to select the appropriate OCI AI service or combination of services based on data type and requirements.
OCI Data Science Platform
- Describe OCI Data Science: managed JupyterLab notebooks, model catalog, model deployment endpoints, and jobs for batch processing.
- Explain OCI Data Labeling for creating annotated training datasets for classification and object detection custom models.
- Describe OCI Data Flow for Apache Spark applications supporting large-scale data processing and feature engineering pipelines.
- Explain model deployment options: real-time HTTP endpoints, model versioning, and A/B testing for production ML serving on OCI.
- Describe OCI Data Integration service for building ETL pipelines that prepare and transform data for ML model training.
- Identify accelerated data science (ADS) SDK capabilities for simplifying model training, evaluation, and deployment workflows.
- Compare pre-built AI services versus custom Data Science models for different complexity and customization requirements.
5
OCI Generative AI Service
2 topics
OCI GenAI Features and Models
- Describe OCI Generative AI Service: text generation, summarization, embedding endpoints, and dedicated GPU cluster hosting.
- Identify foundation models in OCI GenAI including Cohere Command, Cohere Embed, and Meta Llama models and their capabilities.
- Explain OCI GenAI playground for testing generation parameters including temperature, top-p, top-k, max tokens, and stop sequences.
- Describe OCI GenAI fine-tuning for customizing models with enterprise data while maintaining data isolation and security.
OCI GenAI Integration and Architecture
- Explain OCI GenAI embedding endpoint for creating vector representations used in semantic search and RAG architectures.
- Describe OCI GenAI Agents framework for building conversational AI with tool integration, knowledge bases, and retrieval capabilities.
- Explain how OCI GenAI integrates with Oracle Database 23ai Vector Search for enterprise RAG solutions with structured data.
- Describe OCI GenAI security controls including dedicated endpoints, private networking, data residency, and model isolation.
- Identify OCI GenAI pricing model including compute unit consumption, token-based billing, and dedicated cluster cost structures.
- Analyze a generative AI deployment to recommend OCI GenAI models, parameters, guardrails, and integration architecture.
Scope
Included Topics
- All domains in the OCI AI Foundations Associate (1Z0-1122) exam: AI/ML Concepts, Deep Learning, Generative AI, OCI AI Services, OCI Generative AI Service.
- Foundational AI/ML concepts: supervised/unsupervised learning, neural networks, NLP, computer vision, clustering, regression, classification, evaluation metrics, responsible AI.
- OCI AI services: Vision, Speech, Language, Document Understanding, Anomaly Detection, Data Science, Data Labeling, and OCI Generative AI Service.
- Generative AI concepts: LLMs, transformers, prompt engineering, RAG, fine-tuning, embeddings, and responsible AI guardrails.
Not Covered
- Advanced model training, hyperparameter optimization, and custom architecture design beyond foundational understanding.
- Python programming, ML framework code, and SDK-level implementation of AI pipelines.
- Production MLOps, CI/CD for ML, and model serving infrastructure beyond basic awareness.
- Non-Oracle AI services, academic research methods, and specialized domain-specific AI applications.
Official Exam Page
Learn more at Oracle
1Z0-1122 is coming soon
Adaptive learning that maps your knowledge and closes your gaps.
Create Free Account to Be Notified