abc.pdf

Unit 1 AI Reflection , Project Cycle , Ethics Subject: AI 417 Grade: IX _______________________________________________________________________________ Ar...

Unit 1 AI Reflection , Project Cycle , Ethics Subject: AI 417 Grade: IX _______________________________________________________________________________ Artificial Intelligence Artificial Intelligence has always been a term which intrigues people all over the world. Artificial intelligence (abbreviated: Al or A.I.) is how Google ranks pages, Amazon knows what we like, chatbots like Siri chat, and computers play Chess and Go. Artificial intelligence or Al refers to software technologies that make a robot or computer act and think like a human. Artificial intelligence (AI) is the simulation of human intelligence processes by machines, especially computer systems. Speech recognition, decision-making, visual perception, language translation are some features of human intelligence that artificial intelligence may possess. Artificial Intelligence Subcategories The two most refined versions of artificial intelligence are Narrow Al and General Al. Narrow Al Narrow Al is designed to perform tasks that are more specific. It is also called Weak AI. This implies that Narrow Al are intelligent systems that are programmed to perform specific tasks without the need for intrinsic coding. For instance, interpreting audio-visual feeds, organising personal or business schedule, automated customer service assistance, etc. fall under the category of Narrow AI. Virtual personal assistants, such as Apple's Siri, are a form of weak Al. General Al Artificial general intelligence, unlike Narrow Al, includes the capability of understanding a vast scope of activities. It is also called Strong AI. This AI is looked upon as a form of human intelligence as is shown in many popular movies like Ex Machina, The Terminator, and 2001: A Space Odyssey. It is an Al system with generalised human cognitive abilities. When presented with an unfamiliar task, a strong Al system is able to find a solution without human intervention. Depending on the type of data, we can divide AI into different domains: Advantages of Artificial Intelligence over Human Intelligence The Al system would have a low error rate compared to humans, if coded properly. They would have incredible precision, accuracy and speed. While a doctor makes a diagnosis in approximately 10 minutes, an Al system can make a million in the same time. The Al system won't be affected by hostile environments, thus would be able to complete dangerous tasks, explore in space, and endure problems that would injure or kill us. This can even mean mining and digging fuels that would otherwise be hostile for humans. The Al system can replace humans in repetitive, tedious tasks and in many laborious places of work. The Al system can predict what a user will type, ask, search, and do. They can easily act as assistants and can recommend or direct various actions. An example of this can be your own smartphone. The AI system can detect fraud in card-based systems, and possibly other systems in the future. It can also organise and manage huge volume of data. Disadvantages of Artificial Intelligence Humans can become too dependent on Al and lose their mental capacities as seen partially with smartphones and other technology already. Machines can easily lead to destruction, if put in the wrong hands. Al as robots can supercede humans, enslaving us.  Can cost a lot of money and time to build, rebuild and repair. Robotic repair can be used to reduce time and humans needing to fix it, but that'll cost more money and resources. It's questionable: is it ethically and morally correct to have androids, human-like robots, or recreate intelligence, a gift of nature that shouldn't be recreated? Storage is expensive, but access and retrieval may not lead to connections in memory as well as humans could. They can learn and get better with tasks if coded to, but it's questionable as to if this can ever become as good as humans can do They cannot work outside of what they were programmed for. They could never, or, at least, seemingly never with our technological perceptions, receive creativity that humans have. Robots, can lead to severe unemployment by replacing their jobs. No emotions are present in an Al system. Even if coded with common sense and to learn, it seems hard for them to get as much common sense that humans could. Applications and Future Trends of Artificial Intelligence EVERYDAY USES OF AI Siri It is a popular personal assistant offered by Apple. The friendly female voice-activated assistant interacts with the user on a daily routine. She assists us to find information, get directions, send messages, make voice calls, open applications and add events to the calendar, etc. Netflix It is a widely popular content-on-demand service that uses predictive technology to offer recommendations on the basis of consumers' reaction, interests, choices and behavior. The technology examines from a number of records to recommend movies based on your previous liking and reactions. Music and Media Whether you use something like Spotify or enjoy watching Netflix or even YouTube, artificial intelligence is helping you find the music and media that you want. Over time, it learns based on your selections and then provides recommendations for you to add to your playlist. Smartphones The use of artificial intelligence is used with the photo editor on smartphones. When you want to take a picture, artificial intelligence helps by selecting the appropriate settings and suggesting different modes to you. Smart Home Devices Artificial intelligence is used in smart home devices to adjust the temperature and lighting based on our preferences. Online Services From travel to banking, shopping, education and entertainment, these industries rely heavily on artificial intelligence for using chatbots or through algorithms that enable it to track spending, suggest purchases, prevent fraud and complete other transactions much faster.  ARTIFICIAL INTELLIGENCE PROJECT CYCLE In earlier classes we have studied about Water Cycle, etc. In water Cycle we studied that how is the journey of water carried out from one step to other till the end. Like that in Project Cycle we are going to deal with the steps involved in creating a project, starting from the given problem till the project is created and tested. Project Cycle is a step by step process to solve the problems using proven scientific methods and drawing the inference about it. Why do we need an AI Project Cycle? AI Project Cycle – Defined! AI project cycle is the cyclical process followed to complete an AI project. AI project cycle takes us through different steps involved in a project. AI project cycle helps us: ▪ to create better AI projects easily ▪ to create AI projects faster ▪ to understand the process Stage 1 : Problem Scoping Problem Scoping refers to understanding a problem finding out various factors which affect the problem, define the goal or aim of the project. The 4W's of Problem Scoping are Who, What, Where and Why. These Ws helps in identifying and understanding the problem in a better and efficient manner. 1. Who - "Who" part helps us in comprehending and categorizing who all are affected directly and indirectly with the problem and who are called the Stake Holders 2. What - "What" part helps us in understanding and identifying the nature of the problem and under this block, you also gather evidence to prove that the problem you have selected actually exists. 3. Where - "Where" does the problem arises, situation and the location. 4. Why - "Why" is the given problem worth solving. 4W canvas problem statement Example 1 ‘Toppers’ is an educational institution for preparing IIT/JEE examinations. Students of the institute are not happy with the administration. They frequently experience difficulties in receiving important alerts on time like information regarding campus interviews, training and placement drives, vacations, and other special announcements. The institution is finding it difficult to interact with its students and staff after working hours or on weekends. Delays in communication cause a loss of good opportunities and interest in students. For the situation given above create a 4W Project Canvas. Example 2 Create a 4W Project Canvas for the following. As more and more new technologies get into play, risks will get more concentrated into a common network. Cybersecurity becomes extremely complicated in such scenarios and goes beyond the control of firewalls. It will not be able to detect unusual activity and patterns including the movement of data. Think how AI algorithms can scrape through vast amounts of logs to identify susceptible user behavior. Use an AI project cycle to clearly identify the scope, how you will collect data, model and evaluation parameters. Example 3 The world’s largest diamond, is in danger as Mr. X has threatened to steal it. No one is able to track Mr. X and so the situation is critical. You have been appointed as the Chief Security Officer and your job is to enhance the security of the diamond to make the area impossible for Mr X to break into and steal the diamond. Now that you are aware of AI concepts, plan to use them in accomplishing your task. Start with listing down all the factors which you need to consider while framing a security system. Data Acquisition is he process of collecting accurate and reliable data to work with. Data Can be in the format of text, video, images, audio and so on and it can be collected from carious source like the interest, journals, newspapers and so on. For AI project to be efficient, the training data should be authentic and relevant to the problem statement scoped. Training data vs Testing data There are two key types of data used for machine learning training and testing data.  Machine learning algorithms learn from data in datasets. They find patterns in the data, develop an understanding of the data, make decisions based on the data, and evaluate the accuracy of their choices. In machine learning, datasets are typically split into two subsets: training and testing data. The training data is used to train the machine learning algorithm. The testing data is used to evaluate the accuracy of the trained algorithm. The training data should represent the data the algorithm will encounter in the real world. For any AI project to be efficient, the training data should be authentic and relevant to the problem statement scoped. Data Features Data features refer to the type of data you want to collect. Data plays an important role in an AI project as it creates the base on which the AI project is built. Therefore, the acquired data should be authentic, reliable and correct. Also, the acquisition methods should be authentic so that our project is not in conflict with anyone. There can be various ways in which we can collect data. Some of them are : 1. Surveys 2. Web scraping 3. Sensors 4. Cameras 5. Observations 6. APIs Data Exploration is the process of arranging the gathered data uniformly for a better understanding. Data can be arranged in the form of a table, plotting a chart or making database. If we simplify this Data Exploration means that the data which we collected in Data Acquisition, in Data Exploration we need to arrange it for example if we have data of 50 students in a class, we have their Mobile Number, Date of Birth, Class, Etc. In the process of data exploration we can make a chart for that data in which all the names will be at one place and all the mobile numbers at one etc.  Data Exploration or Visualization Tools Stage 4 : Modelling AI Modelling refers to developing algorithms, also called models which can be trained to get intelligent outputs. That is, writing codes to make a machine artificially intelligent. Rule Based Approach Rule Based Approach Refers to the AI modelling where the relationship or patterns in data are defined by the developer. The machine follows the rules or instructions mentioned by the developer, and performs its task accordingly. Decision Tree It is a rule-based AI model which helps the machine in predicting what an element is with the help of various decisions (or rules) fed to it. A basic structure of decision tree is shown below: Here, the Decision tree starts from the question Am I Hungry? The beginning point of any Decision Tree is known as its Root. It then forks into two different ways or conditions: Yes or No. The forks or diversions are known as Branches of the tree. The branches either lead to another question, or they lead to a decision like Go to Sleep which is known as the leaf. If you look closely at the image above, you will notice that it looks like an inverted tree with root above and the leaves below. Hence the name Decision Tree! Learning Based Approach Refers to the AI modelling where the machine learns by itself. In this approach, random data is fed to the machine and it is left on the machine to figure out patterns and trends out of it.  Thus, the machine looks at the data, tries to extract similar features out of it and clusters same datasets together. In the end as output, the machine tells us about the trends which it observed in the training data.  Artificial Intelligence, or AI, refers to any technique that enables computers to mimic human intelligence. The AI-enabled machines think algorithmically and execute what they have been asked for intelligently. Machine Learning, or ML, enables machines to improve at tasks with experience. The machine learns from its mistakes and takes them into consideration in the next execution. It improvises itself using its own experiences. Deep Learning, or DL, enables software to train itself to perform tasks with vast amounts of data. In deep learning, the machine is trained with huge amounts of data which helps it into training itself around the data. Such machines are intelligent enough to develop algorithms for themselves. Deep Learning is the most advanced form of Artificial Intelligence out of these three. Then comes Machine Learning which is intermediately intelligent and Artificial Intelligence covers all the concepts and algorithms which, in some way or the other mimic human intelligence. Learning Based Approach Refers to the AI modelling where the machine learns by itself. Under the Learning Based approach, the AI model gets trained on the data fed to it and then can design a model which is adaptive to the change in data. Supervised Learning In a supervised learning model, the dataset which is fed to the machine is labelled. In other words, we can say that the dataset is known to the person who is training the machine only then he/she is able to label the data. A label is some information which can be used as a tag for data. There are two types of Supervised Learning models: Classification and Regression Classification Where the data is classified according to the labels. For example, in the grading system, students are classified on the basis of the grades they obtain with respect to their marks in the examination. This model works on discrete dataset which means the data need not be continuous. Classification problems ask the algorithm to predict a discrete value, identifying the input data as a member of a particular class, or group. In a training dataset of animal images, that would mean each photo was pre- labeled as Cat or Dog. The algorithm is then evaluated by how accurately it can correctly classify new images of other cats or dogs. In classification, the algorithm is able to determine which set a given data point belongs to by means of a classification function. The model classifies datasets according to the rules given to it. Usually the dataset used for classification are labelled and the data then gets sorted according to their labelling. Testing data (New Data) is then classified as one of the labels of the training dataset (Old data). For example: If we want to train a model to identify if an image is of an Apple or Cherry, we need to train it with multiple images of both Apple and Cherry along with their labels. The machine will then classify images on the basis of the labels and predict the correct label for Test data. Regression In Regression, the algorithm generates a mapping function from the given data. With the help of this mapping function, we can predict the future data. For example, if we want to predict the salary of an employee, we can use his past salaries as training data and can predict his next salary. Supervised learning is, thus, best suited to the problems where there is a set of available reference points or a ground truth with which to train the algorithm. Differences between Classification and Regression Regression and Classification algorithms are Supervised Learning algorithms. Both the algorithms are used for prediction in Machine learning and work with the labeled datasets.  But the difference between both is how they are used for different machine learning problems. The main difference between Regression and Classification algorithms that Regression algorithms are used to predict the continuous values such as price, salary, age, etc. and Classification algorithms are used to predict/Classify the discrete values such as Male or Female, True or False, Spam or Not Spam, etc. Unsupervised Learning An unsupervised learning model works on unlabeled dataset. This means that the data which is fed to the machine is random and there is a possibility that the person who is training the model does not have any information regarding it. The unsupervised learning models are used to identify relationships, patterns and trends out of the data which is fed into it. It helps the user in understanding what the data is about and what are the major features identified by the machine in it. For example, you have a random data of 1000 dog images and you wish to understand some pattern out of it, you would feed this data into the unsupervised learning model and would train the machine on it. After training, the machine would come up with patterns which it was able to identify out of it. The Machine might come up with patterns which are already known to the user like colour or it might even come up with something very unusual like the size of the dogs. Unsupervised learning models can be further divided into two categories: Clustering Dimensionality Reduction Clustering Refers to the unsupervised learning algorithm which can cluster the unknown data according to the patterns or trends identified out of it. The patterns observed might be the ones which are known to the developer, or it might even produce some unique patterns out of it.  Stage 5 : Evaluation Once a model has been made and trained, it needs to go through proper testing so that one can calculate the efficiency and performance of the model. Hence, the model is tested with the help of Testing Data (which was separated out of the acquired dataset at Data Acquisition stage) and the efficiency of the model is calculated on the parameters mentioned below: There are various new terminologies which come into the picture when we work on evaluating our model. Stage 6 : Deployment What is deployment? Deployment as the final stage in the AI project cycle where the AI model or solution is implemented in a real-world scenario. Key Steps in Deployment Process the key steps involved in the deployment process: a. Testing and validation of the AI model b. Integration of the model with existing systems c. Monitoring and maintenance of the deployed model. Some examples of successful AI projects that have been deployed in various industries, such as selfdriving cars, medical diagnosis systems, and chatbot. Ethics and Morality What are the principles of AI Ethics? AI Ethics Principles Identifying the principles To make AI better, we need to identify the factors responsible for it. The following principles in AI Ethics affect the quality of AI solutions ▪ Human Rights ▪ Bias ▪ Privacy ▪ Inclusion Let’s look at the AI Ethics principles in detail: Human Rights When building AI solutions, we need to ensure that they follow human rights. Here are a few things that you should take care of ▪ Does your AI take away Freedom? ▪ Does your AI discriminate against People? ▪ Does your AI deprive people of jobs? ▪ What are some other human rights which need to be protected when it comes to AI? Bias Bias (partiality or preference for one over the other) often comes from the collected data. The bias in training data also appears in the results. Here are a few things that you should take care of ▪ Does your data equally represent all the sections of the included populations? ▪ Will your AI learn to discriminate against certain groups of people? ▪ Does your AI exclude some people? ▪ What are some other biases that might appear in an AI? Privacy We need to have rules which keep our individual and private data safe. Here are a few things that you should take care of ▪ Does your AI collect personal data from people? ▪ What does it do with the data? ▪ Does your AI let people know about the data that it is collecting for its use? ▪ Will your AI ensure a person’s safety? Or will it compromise it? ▪ What are some other ways in which AI can breach someone’s privacy? Inclusion AI MUST NOT discriminate against a particular group of population, causing them any kind of disadvantage. Here are a few things you should take care of ▪ Does your AI leave out any person or a group? ▪ Is a rich person and a poor person benefitted equally from your AI? ▪ How easy is it to use your AI? ▪ Who does your AI help? ▪ How can we make AI more inclusive?