Page 352 - Ai_417_V3.0_C9_Flipbook
P. 352

3.  List the important features of real images captured by camera.
                Ans.  Real images captured by cameras are visual representations depicting scenes, objects, or people in exactly the same
                    way as they exist in the real world. These images are created by humans or nature and are unaltered.
                  4.  What is a supervised learning model?
                Ans.  In a supervised learning model, a labelled dataset is given to the machine. A labelled dataset is the information, which
                    is tagged with identifiers of data. For example, clothes in a store are marked under various categories of clothing like
                    Shirts, Trousers, Coats, etc. They are further labelled as per gender and size.
                  5.  Explain the term deepfake.

                Ans.  The term "deepfake" combines "deep learning" and "fake," referring to AI techniques that create realistic but fake videos
                    and audio. These AI-generated videos can mislead people by making it seem like someone said or did something they
                    didn't, undermining trust in public figures and institutions.

                  6.  What is the use of Generative Modelling?
                Ans.  Generative modelling is a specific approach within unsupervised learning that focuses on understanding and modelling
                    how the data is generated.

                  7.  What is a random noise dataset?
                Ans.  A "random noise dataset" typically refers to a collection of data points or samples where each data point is generated
                    randomly. There are unpredictable fluctuations and disorganised data, which makes it impossible to identify target
                    patterns or relationships in it. This may result in decreased accuracy or reliability of the output, which the generative AI
                    model takes care of.

                  8.  What is Generative AI?
                Ans.  Generative Artificial Intelligence (AI), also called Gen AI, refers to the algorithms that generate new data that resembles
                    human-generated content, such as audio, code, images, text, simulations, and videos.
                  9.  Differentiate between Generative AI and Conventional AI in terms of goals.

                Ans.  Generative AI creates new content that mimics the original content. This content includes images, text, music, or other
                    forms of media. Whereas conventional AI analyses, processes, and classifies data. It basically works to improve the
                    accuracy, precision, recall, and speed within the scope of the defined task.

                 10.  Explain VAEs with example.
                Ans.  This is another class of generative models. In order to produce fresh data, VAEs learn the distribution of the data and
                    then sample from it.

                    Example: Generation of new images similar to the given training set. For instance, a VAE trained on images of faces can
                    generate new, realistic-looking faces.


              B.  Long answer type questions:
                  1.  How does Discriminative Modelling contrast with Generative Modelling?
                Ans.  In supervised learning, discriminative modelling contrasts with generative modelling, where the goal is to model the
                    joint probability distribution of both the input features and the output labels. Generative models  can be used to
                    generate new data points that resemble the training data, whereas discriminative models are primarily focused on
                    classification or regression tasks.

                  2.  What are the limitations of Gen AI?
                Ans.  Limitations of Using Generative AI
                    ●   Data bias: If generative AI is trained on biased or incomplete data, the output may be similarly biased or flawed.
                       This can lead to inaccurate or problematic results in certain applications, such as in facial recognition or natural
                       language processing.


                    350     Touchpad Artificial Intelligence (Ver. 3.0)-IX
   347   348   349   350   351   352   353   354   355   356   357