Artificial Intelligence(AI)

AI is the massive goal. It's the overall field of making computers smart, giving the machines the ability to mimic human intelligence. A computer system that can act like human like learn, perceive, reasoning problems solving.

Machine Learning(ML)

ML is the subset of Artificial Intelligence where systems learn from data to improve performance. ML is the process of teaching computers to learn patterns from data and make decisions based on other patterns. It uses algorithms to find patterns , large datasets, create models, making decisions. Examples includes Data(Accept or Reject), fraud detection, Spam.

Types of Machine Learning(ML)

Supervised Learning: models learn from labeled data(data in which input-output exists in a structured manner), trains algorithms using datasets to predict correct outcomes for new or unseen data.

Classification: Assigning data points to specific categories. For example, classifying emails as "spam" or "not spam," or identifying handwritten digits.

Regression: Predicting a continuous numerical value. For example, predicting house prices based on features like size and location, or forecasting sales revenue.

Unsupervised Learning: Models learn from unlabeled data(data in which input and output not exists).The algorithm works with data that has not been pre-classified or labeled.

Clustering: Groups data points into clusters based on their similarities. For example, grouping customers with similar purchasing behaviors.

Dimensionality reduction: Simplifies data by reducing the number of features while retaining important information. This makes the data easier to work with.

Association rule: Identifies relationships between items in large datasets, such as finding items that are frequently purchased together in a store.

Reinforcement Learning

Reinforcement Learning revolves around the idea that an agent (the learner or decision-maker) interacts with an environment to achieve a goal. The agent performs actions and receives feedback to optimize its decision-making over time.

Some RL Algorithms:

Q-Learning

Deep Q networks(DQN)

Policy Gradient methods

Proximal policy optimization(PPO)

Tools used: Python, Jupyter, NumPy, Pandas, Seaborn, Scikit, MatplotLib, XGBoost.

Deep Learning:

Uses advanced computing, it's own networks to adapt with little to no human intervention.

A powerful subset of machine learning which uses artificial neural networks inspired by Human brain to automatically learn complex patterns from large amount of data for making predictions and giving results.

Tools used in neural networks :

TensorFlow(made by Google)

PyTorch(made by meta)

Kaggle

Key Applications:

Computer Vision

Recommendation system

Natural language processing

Speech recognition

Neural Network Architectures:

FNN(Feed Forward Network): FNN is a type of artificial neural network where information flows in one direction from input to output without loops used in machine learning for tasks like image recognition and data classifications. Once we give data it passes through multiple layers and give predictions. These type of neural networks not used for sequential data and time dependent data.

RNN(Recurrent Neural Network): RNN is a type of artificial neural networks where information flows in back direction to use the data. These type of neural networks allows to process sequential data.

CNN(Convolutional Neural Network): A CNN, or convolutional neural network, is a type of neural network designed for processing grid-like data, most commonly images.

Transformers: Transformers are powerful neural network architectures, introduced in 2017, that use a self-attention mechanism to weigh the importance of different parts of an input sequence, enabling parallel processing and capturing long-range dependencies, unlike older models like RNNs like GPT(Generative Pre-defined transformers)

Gen AI(Generative Artificial Intelligence): Gen AI is a type of AI that used to create original content, images, audios, code, videos learning from massive datasets. Using deep learning and neural networks to produce outputs often indistinguishable from human-made work, enabling tasks like content creation, summarization, research, and personalized assistance across many industries.

Key Applications:

Content creation

Software Development

Research and Analysis

Education

Customer Services

Computer Vision: It uses deep learning, neural networks, and algorithms to process visual data, powering applications from self-driving cars and facial recognition to medical diagnostics and manufacturing quality control. Very Popular is CNNs.

Example: face recognition, self driving cars.