What you will learn in The Complete Obsolete Guide to Generative AI
Chapter by chapter
Concepts explained in this book
- language modelA machine-learning model trained on large text corpora to process and generate natural-language text.
- conversational aiSoftware that interacts with users conversationally and performs information, control, or task-oriented operations.
- promptA prompt is an input instruction or context supplied to a generative artificial-intelligence model to elicit an output.
- GPT modelA family of generative language models based on the Generative Pre-trained Transformer architecture.
- ChatPDFChatPDF is a web application that analyzes PDF documents through a conversational artificial-intelligence interface.
- datasetA dataset is a structured collection of data used to analyze patterns or train a computational model.
- artificial intelligenceThe field concerned with computational systems that perform tasks associated with human intelligence, such as generating language or making inferences.
- LlamaIndexLlamaIndex is a software framework for building large-language-model applications that connect models with external or private data.
- ChatPDF APIThe ChatPDF API is a programmatic interface for uploading PDF documents, creating document-query sessions, and requesting answers about them.
- speech recognitionSpeech recognition is the conversion of spoken audio into a machine-readable representation of language.
- automationAutomation is the execution of tasks or processes by software or machines with little or no manual intervention.
- sentiment analysisSentiment analysis is the computational classification of text according to expressed affective polarity such as positive, negative, or neutral.
- prompt engineeringPrompt engineering is the deliberate design and refinement of instructions and context supplied to a generative model to influence the quality of its output.
- investingAn investment decision is a choice about allocating capital among securities or other investment assets.
- ChatGPTA version of the ChatGPT conversational artificial-intelligence service released in 2022.
- AutoGPTAuto-GPT is an open-source autonomous-agent application that uses large language models to decompose goals and perform multi-step tasks.
- normalizationData normalization is the transformation of values to a comparable scale or representation so differences among variables can be meaningfully visualized or analyzed.
- GPUA massively parallel processor originally designed for graphics workloads and widely used for numerical machine-learning computation.
- data analysisData analysis is the examination and transformation of data to discover values, patterns, or other useful findings.
- few-shot learningFew-shot learning is adapting a model to a task by conditioning it on a small number of task-specific examples or demonstrations.
- business analyticsThe analysis of business data to identify patterns, relationships, and insights that support business decisions.
- data wranglingThe process of transforming, cleaning, and structuring raw data into a consistent machine-readable form.
- statistical relationshipA statistical relationship is an association between variables supported by quantitative analysis of observed data.
- generative modelingA machine-learning model that produces new content resembling patterns learned from its training data.
- CSVCSV is a text-based tabular data format in which fields are commonly separated by commas and records are separated by lines.
- transparencyTransparency is the availability of information about a system's design, operation, decision-making, and effects sufficient for external understanding or scrutiny.
- shell scriptBash is a command-line shell and scripting language for interacting with operating systems and executing shell programs.
- AI agentAn AI agent is a software system that uses artificial-intelligence models to pursue goals by selecting and carrying out actions.
- PythonPython is a general-purpose programming language commonly used for scripting, automation, and software development.
- narrow aiNarrow AI is artificial intelligence designed to perform specific tasks or operate within specific domains rather than general intellectual activities.
- LangChainLangChain is a software framework for building applications that coordinate large language models with tools, data sources, and multistep operations.
- curlcurl is an open-source command-line tool for transferring data through network protocols.
- fine-tuningFine-tuning is continued training of a pretrained model on task- or domain-specific data to specialize its behavior.
- quantizationA quantized model is a machine-learning model whose numerical parameters or activations are represented at reduced precision to lower resource requirements.
- biasBias is a systematic tendency for judgments or decisions to favor or disadvantage people for reasons unrelated to relevant qualifications.
- low-rank adaptationLoRA is a parameter-efficient fine-tuning technique that adapts a pretrained model by learning low-rank updates rather than modifying all model parameters.
- GPTSimpleVectorIndexGPTVectorStoreIndex is a LlamaIndex index component that represents source data through embeddings for language-model retrieval.
- parameterA model parameter is a learned numerical value that determines part of a machine-learning model's behavior.
- parallelismParallel computation is the simultaneous execution of multiple computational operations.
- debuggingTroubleshooting is the systematic diagnosis and correction of faults in a system or process.
- temperatureTemperature is a generation parameter that controls the randomness or diversity of a language model's output.
- domain knowledgeSpecialized understanding and expertise concerning a particular subject or professional field.
- 10-k filingA Form 10-K is an annual regulatory filing that publicly traded companies submit to the U.S. Securities and Exchange Commission with detailed financial and business information.
- Portfolio PilotPortfolio Pilot is an investment-analysis service that applies artificial-intelligence models to market conditions and individual portfolio information.
- due diligenceDue diligence is the investigation and verification of relevant information before making a financial or business decision.
- investment researchInvestment research is the systematic analysis of companies, securities, markets, and financial information to support investment decisions.
- min-max normalizationMin-max scaling is a normalization method that linearly transforms values into a specified range, commonly from 0 to 1.
- frequency penaltyA generation parameter that penalizes tokens in proportion to how often they have already been sampled, reducing repeated words or phrases.
- trainingModel training is the process of fitting a model by exposing it to data so it learns patterns for making predictions or generating outputs.
- market volatilityA measure of the magnitude and frequency of price fluctuations in a financial market or asset.
- retail investorAn individual investor who participates in financial markets outside the role of an institutional investment organization.
- systemic riskThe risk that disruption in one part of a financial system propagates broadly enough to impair the system as a whole.
- OpenAI PlaygroundA web-based interface for experimenting with OpenAI language models, prompts, and generation settings.
- attentionAn attention mechanism is a neural-network component that weights relationships among input elements when computing representations or outputs.
- model architectureThe structural design of a machine-learning model, including its components and their organization.
- public domainThe status of creative work that is not protected by copyright or whose copyright restrictions have expired or been waived.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.