Page 312 - Information_Practice_Fliipbook_Class11
P. 312

AI tools also personalize our learning by enabling us to focus on our weak points. Another  AI tool may help a bank
        detect fraudulent transactions. Barring an occasional error, AI technology successfully filters out spam emails from
        our mailboxes. Thus, we note that AI covers several aspects of our lives, such as  healthcare, finance, transportation,
        education, and entertainment.
        However, AI also raises ethical and societal concerns, such as the displacement of people from jobs, privacy, and safety.
        It is important for AI developers and researchers to address these concerns and design AI systems that are ethical,
        transparent, and accountable.


               Artificial intelligence (AI) is a branch of computer science that aims to make machines so intelligent that they can
               perform tasks that typically require human intelligence.


        10.1.1 Machine Learning

        Machine learning (ML) is a subdomain of artificial intelligence (AI) that involves the use of statistical and specialized
        algorithms to enable computer systems to learn and improve from experience without being explicitly programmed.
        In other words, machine learning systems use data to learn and make decisions rather than relying on pre-determined
        rules or instructions.

                                            Machine Learning Process










                        Training Data    Algorithm       Learning     Trained model      Results

                                                  Machine Learning Process
        There are three main approaches to machine learning: supervised learning, unsupervised learning, and reinforcement
        learning. In the machine learning community, a learner is called a model or an agent.

        Supervised learning involves training a model on labelled data to make predictions or classifications based on new,
        unlabelled data. In supervised learning, the goal is to learn a function that maps input data to output data based on
        example input-output pairs. The labelled data used to train the model consists of input-output pairs, where the input is
        a feature vector and the output is a label or target variable. The model learns from these input-output pairs by finding
        a function that approximates the relationship between the input and output variables. This function can then be used
        to make predictions or classifications on new, unlabelled data. For example, suppose we have a dataset of images of
        handwritten digits, where each image is labelled with the corresponding digit (0-9). We can use this dataset to train
        a supervised learning model to recognize digits in new, unseen images. We might use an algorithm such as logistic
        regression or a neural network to learn a mapping between the input images and their corresponding labels. During
        the training phase, the model adjusts its parameters to minimize the difference between its predicted output and the
        true output labels. This is typically done using an optimization algorithm such as gradient descent. Once the model has
        been trained, it can be used to predict the labels of new, unseen images by applying the learned function to the input
        features of those images.
        Unsupervised  learning  involves  training  a  model  on  unlabelled  data  to  identify  patterns  or  relationships  in  the
        data.  Reinforcement  learning  involves  training  a  model  to  make  decisions  based  on  feedback  received  from  the
        environment.





          298  Touchpad Informatics Practices-XI
   307   308   309   310   311   312   313   314   315   316   317