Subject: Artificial intelligence (CSC 309)
Artificial Intelligence (AI) is the field of computer science that focuses on creating machines or software capable of performing tasks that normally require human intelligence (i.e machines that can mimic human intelligence). These tasks include learning, reasoning, problem-solving, perception, language understanding, and decision-making.
AI systems work by using: - Data
- Algorithms
- Computational power
to mimic or enhance human capabilities.
History of AI
1. Early Ideas (Before 1950s) - Ancient myths of artificial beings.
- Mathematicians like Leibniz and Boole developed logic foundations.
- Alan Turing (1950) proposed the Turing Test for intelligent behavior.
2. Birth of AI (1956) - Term “Artificial Intelligence” coined at Dartmouth Conference by John McCarthy.
- Early programs could play checkers or solve algebra.
3. The Golden Years (1956–1974) - Development of problem-solving programs.
- Introduction of symbolic AI.
4. AI Winters (1974–1980, 1987–1993) - Funding dropped due to slow progress.
5. Expert Systems (1980s) - AI used in industries to imitate human experts.
- Started commercial growth.
6. Rise of Machine Learning (1990s–2010) - Statistical methods, data-driven models.
- Advancements in speech recognition and computer vision.
7. Modern AI (2010–Present) - Deep learning breakthroughs.
- AI applications in autonomous cars, robotics, healthcare, and generative AI (ChatGPT, Midjourney, etc.).
Goals of AI
1. Automation: Perform tasks without human intervention.
2. Reasoning: Make logical decisions.
3. Learning: Improve performance from data and experience.
4. Understanding: Interpret language, images, sounds.
5. Problem-Solving: Solve complex tasks better or faster than humans.
6. Perception: Sense the environment like humans do.
7. Creativity: Generate new ideas, art, music, and designs.
Types of AI
1. Weak/Narrow AI: Specialized systems, designed for one specific task such as chatbots and recommendation engines. Examples: Siri, ChatGPT, Google Search, Facial recognition.
2. Strong/General AI: Hypothetical systems (still theoretical) with human-level intelligence across domains (not limited to one specific task only). Can learn any task a human can.
3. Superintelligent AI: Future concept where AI surpasses human intelligence. Hypothetical future system
Techniques in AI
1. Supervised Learning: Training models with labeled data.
2. Unsupervised Learning: Discovering hidden patterns in unlabeled data.
3. Reinforcement Learning: Learning through trial and error with rewards.
4. Deep Learning: Multi-layered neural networks for complex tasks like speech recognition
Techniques in AI
1. Supervised Learning: Training models with labeled data.
2. Unsupervised Learning: Discovering hidden patterns in unlabeled data.
3. Reinforcement Learning: Learning through trial and error with rewards.
4. Deep Learning: Multi-layered neural networks for complex tasks like speech recognition
Branches of AI
1. Machine Learning: Learning from data.
2. Deep Learning: Neural networks with many layers
3. Natural Language Processing (NLP): Understanding human language.
4. Computer Vision: Understanding images and videos.
5. Expert Systems: rule-based systems acting like human experts.
6. Robotics: Intelligent machines that interact with the physical world.
Applications of AI
1. Healthcare: diagnostics, drug discovery, personalized treatment.
2. Finance: fraud detection, algorithmic trading, customer support.
3. Transportation: self-driving cars, traffic management.
4. E-commerce: personalized recommendations, chatbots.
5. Education: adaptive learning platforms and AI tutors
6. Security: surveillance analytics and cybersecurity threat detection
7. Entertainment: video game AI and movie recommendations
Advantages of AI
1. Accuracy and Efficiency: performs tasks faster and more accurately.
2. Automation: reduces human workload.
3. No Fatigue: it can work continuously without breaks.
4. Data Handling: processes large amounts of information quickly.
5. Decision Support: helps in making informed decisions.
6. Increased Safety: used in dangerous environments (mining, bomb disposal).
Disadvantages of AI
1. Job Loss / Automation: machines can replace human jobs.
2. High Cost: development and maintenance are expensive.
3. Lack of Creativity: AI still struggles with truly original ideas (except within constraints).
4. Dependency: humans may rely too much on AI.
5. No Emotions or Ethics: cannot understand moral nuances.
6. Data Privacy Issues: requires large datasets that may contain personal information.