Revision notes

Information Technology and AI: Foundation

The key points, rules to remember, tables and diagrams from all 5 lessons, on one page for quick revision. To keep a copy, use your browser's Print and choose "Save as PDF".

Lesson 1

Chips and networks: semiconductors, bits and mobile generations from 1G to 5G

What each generation added
GenerationRoughly fromWhat it added
1G1980sMobile voice calls using analog signals; no data, easy to eavesdrop
2G1990sDigital voice (clearer and more secure), SMS text messages and slow data; GSM and CDMA standards
3G2000sPractical mobile internet and video calls
4G2010sFast mobile broadband on an all-IP (internet protocol) network: smooth video streaming and voice over LTE (VoLTE)
5G2020sMuch higher speeds, very low latency and the ability to connect a huge number of devices at once

Byte = 8 bits. Time = size in bits ÷ speed in bits per second. Bandwidth is how much; latency is how late. 1G voice, 2G digital voice and SMS, 3G mobile internet, 4G all-IP broadband, 5G speed + low latency + massive device numbers.

Key takeaways: semiconductors such as silicon can have their conductivity controlled, which lets us build transistors and chips. Data is counted in bits and bytes. A network is judged by bandwidth and latency. Each mobile generation added something new, and 5G is designed for very low delay and very many devices, not only faster phones.

Read the full lesson

Lesson 2

The internet of things, cloud computing and big data

The three service models
Service modelWhat you getWhat you still manageKitchen comparison
IaaS - Infrastructure as a ServiceVirtual servers, storage and networkingOperating system, software and your dataYou hire an equipped kitchen and cook everything yourself
PaaS - Platform as a ServiceA ready platform to build and run your own applicationsOnly your application code and dataKitchen and ingredients are supplied; you follow your own recipe
SaaS - Software as a ServiceFinished software used through a browser or appOnly your own data and settingsYou simply order the finished meal

IaaS = rent the infrastructure; PaaS = rent a platform to run your code; SaaS = use finished software. Big data: volume, velocity, variety (plus veracity and value).

Key takeaways: IoT connects physical objects that sense, send and act. Cloud computing rents computing over the internet, in three service models and four deployment models. The data produced by billions of devices and users is big data, described by its volume, velocity and variety.

Read the full lesson

Lesson 3

Artificial intelligence and how machines learn

Narrow versus general AI
TypeWhat it meansStatus
Narrow (weak) AIBuilt for a specific task or set of tasks - playing chess, recognising faces, recommending videos, translating textAll AI in use today is narrow AI
General AI (AGI)Would match human ability across almost any intellectual task and transfer learning from one area to anotherA goal of research, not an achieved fact
The three main types of machine learning
Type of learningWhat the data looks likeWhat the machine doesEveryday examples
Supervised learningExamples come with the right answer (labels)Learns to predict the label for new examplesSpam filter, recognising handwritten digits, predicting a house price
Unsupervised learningExamples have no labelsFinds hidden groups or patterns by itselfGrouping customers with similar buying habits, spotting unusual transactions
Reinforcement learningNo fixed data set; an agent acts in an environmentLearns by trial and error from rewards and penaltiesGame-playing programs, a robot learning to walk

Supervised = learns from labelled examples. Unsupervised = finds patterns in unlabelled data. Reinforcement = learns by trial, reward and penalty. All AI in use today is narrow AI.

Key takeaways: AI is machines doing tasks that need intelligence; the Turing test (1950) and the term 'artificial intelligence' (1956) mark its beginnings. Today's AI is narrow; general AI remains a research goal. Machine learning learns patterns from data, in three main ways: supervised, unsupervised and reinforcement.

Read the full lesson

Lesson 4

Neural networks, deep learning and generative AI

From the widest idea to the narrowest
TermRelationship
Artificial intelligenceThe broad field of machines doing intelligent tasks
Machine learningA part of AI: learning patterns from data
Deep learningA part of machine learning: neural networks with many layers
Generative AIModels, usually deep learning, that create new content

Deep learning = neural networks with many layers. A language model predicts the next word; fluent is not the same as true. A model is only as fair and accurate as its training data.

Key takeaways: a neural network learns by adjusting the weights between layers of neurons; deep learning uses many layers and needs lots of data and computing power. Generative AI creates new content, and large language models do it by predicting the next word. Their limits - hallucination, bias from training data and misuse such as deepfakes - are as important as their abilities.

Read the full lesson

Lesson 5

Blockchain, quantum computing and India's digital public infrastructure

Three building blocks
SystemPurposeHow it works in brief
AadhaarIdentityA 12-digit number issued by the Unique Identification Authority of India (UIDAI) to residents, linked to biometric details (fingerprints, iris scans, photograph) and basic demographic details; used to confirm 'you are who you say you are'
UPI (Unified Payments Interface)PaymentsBuilt by the National Payments Corporation of India (NPCI); sends money instantly between bank accounts from a mobile phone using a simple UPI ID, across banks and apps
DigiLockerDocumentsA government platform under the Digital India programme where documents such as mark sheets or a driving licence are issued digitally by the issuing authority and can be shared and verified

Blockchain = a ledger copied across many nodes, blocks linked by hashes, new blocks added by consensus. Qubit = superposition of 0 and 1. Aadhaar = identity, UPI = payments, DigiLocker = documents.

Key takeaways: a blockchain makes records tamper-evident by copying them widely and chaining blocks with hashes. Quantum computers use qubits in superposition and entanglement and suit special problems. Aadhaar, UPI and DigiLocker are shared digital building blocks for identity, payments and documents.

Read the full lesson

Privacy · Terms · Refunds · Delivery