interview questions and answers on artificial intelligence

1. What is Artificial Intelligence? Give an example of where AI is used on a daily basis.

“Artificial Intelligence (AI) is an area of computer science that emphasizes the creation of intelligent machines that work and react like humans.” “The capability of a machine to imitate intelligent human behavior.”

Google’s Search Engine
One of the most popular AI Applications is the google search engine. If you open up your chrome browser and start typing something, Google immediately provides recommendations for you to choose from. The logic behind the search engine is Artificial Intelligence.

AI uses predictive analytics, NLP, and Machine Learning to recommend relevant searches to you. These recommendations are based on data that Google collects about you, such as your search history, location, age, etc. Thus, Google makes use of AI, to predict what you might be looking for.

2. What are the different platforms for Artificial Intelligence (AI) development?

Some different software platforms for AI development are-

1. Amazon AI services
2. Tensorflow
3. Google AI services
4. Microsoft Azure AI platform
5. Infosys Nia
6. IBM Watson
7. H2O
8. Polyaxon
9. PredictionIO

3. What Are Intelligent Agents, and How Are They Used in AI?

Intelligent agents are autonomous entities that use sensors to know what is going on, and then use actuators to perform their tasks or goals. They can be simple or complex and can be programmed to learn to accomplish their jobs better. 

4. What are the programming languages used for Artificial Intelligence?

The following are the best AI programming languages used for Artificial Intelligence:

1. Python
2. Java
3. R
4. Prolog
5. Lisp
6. AIML
7. STRIPS
8. Julia.

5. Explain the different domains of Artificial Intelligence.

• Machine Learning: It’s the science of getting computers to act by feeding them data so that they can learn a few tricks on their own, without being explicitly programmed to do so.
• Neural Networks: They are a set of algorithms and techniques, modeled in accordance with the human brain. Neural Networks are designed to solve complex and advanced machine-learning problems.
• Robotics: Robotics is a subset of AI, which includes different branches and applications of robots. These Robots are artificial agents acting in a real-world environment. An AI Robot works by manipulating the objects in its surrounding, by perceiving, moving, and taking relevant actions.
• Expert Systems: An expert system is a computer system that mimics the decision-making ability of a human. It is a computer program that uses artificial intelligence (AI) technologies to simulate the judgment and behavior of a human or an organization that has expert knowledge and experience in a particular field.
• Fuzzy Logic Systems: Fuzzy logic is an approach to computing based on “degrees of truth” rather than the usual “true or false” (1 or 0) boolean logic on which the modern computer is based. Fuzzy logic Systems can take imprecise, distorted, noisy input information.
• Natural Language Processing: Natural Language Processing (NLP) refers to the Artificial Intelligence method that analyses natural human language to derive useful insights in order to solve problems.

6. What is the future of Artificial Intelligence?

Artificial Intelligence has affected many humans and almost every industry, and it is expected to continue to do so. Artificial Intelligence has been the main driver of emerging technologies like the Internet of Things, big data, and robotics. AI can harness the power of a massive amount of data and make an optimal decision in a fraction of a second, which is almost impossible for a normal human. AI is leading areas that are important for mankind such as cancer research, cutting-edge climate change technologies, smart cars, and space exploration. It has taken the center stage of innovation and development of computing, and it is not ceding the stage in the foreseeable future. Artificial Intelligence is going to impact the world more than anything in the history of mankind.

7. How are Artificial Intelligence and Machine Learning related?

Artificial Intelligence and Machine Learning are two popular and often misunderstood words. Artificial Intelligence is a domain of computer science that enables machines to mimic human intelligence and behavior. On the other hand, Machine Learning is a subset of Artificial Intelligence and is all about feeding computers with data so that they can learn on their own from all the patterns and models. Machine Learning models are used to implement Artificial Intelligence frequently. 

AI can be approached in a variety of ways, for as by building a computer program that implements a set of domain expert-developed rules. Artificial Intelligence (AI) incorporates Machine Learning (ML). Machine learning (ML) is the study of creating and implementing algorithms that can learn from previous experiences. If a pattern of behavior has been seen in the past, you can anticipate whether or not it will occur again.

For instance, if you want to develop a program that can identify the animal by looking at the image, you have to use a machine-learning algorithm that can predict the animal in the image based on millions of images stored in the database. The algorithm goes through all the images and classifies each image based on its features (color of pixels, for instance).

8. What Are Neural Networks, and How Do They Relate to AI?

Neural networks are a class of machine learning algorithms. The neuron part of the neural is the computational component, and the network part is how the neurons are connected. Neural networks pass data among themselves, gathering more and more meaning as the data moves along. Because the networks are interconnected, more complex data can be processed more efficiently.

9. How many Types of Artificial Intelligence are there? What are they?

There are four types of Artificial Intelligence as follows:

1. Reactive Machines AI
2. Limited Memory AI
3. Theory of Mind AI
4. Self Aware AI.

10. What are the misconceptions about Artificial Intelligence?

Some misconceptions about Artificial Intelligence that exists are:

• Machines learn from themselves- The reality is far from the statement. Machines are not yet at that stage where they can make a decision on their own. Machines learn through a process called machine learning that enables systems to learn and develop based on their experiences without having to be explicitly programmed. Machine learning is concerned with the creation of computer programs that can access data and learn on their own.
• Artificial Intelligence is the same thing as Machine learning- Artificial Intelligence and Machine learning differ from each other. Artificial Intelligence concerns itself with creating devices that can mimic human intelligence, while machine learning is a subset of artificial intelligence which is about creating programs that can analyze data, learn from it, and then make decisions. 
• Artificial Intelligence will take over humans- There is a possibility that the capabilities of AI can match or even surpass human intelligence in the near future. But, saying that AI will take over humans is just a work of fiction. AI is supposed to complement human intelligence, not enslave it.

11. What are the stages of learning AI?

The following are the stages of learning AI:

• Artificial General Intelligence (AGI): It is also known as Strong AI, which is considered a threat to many scientists’ human existence. It is an evolution of AI where machines can think and make decisions just like humans.
• Artificial Normal Intelligence (ANI): It is also known as Weak AI that can perform only a defined activity set. It does not perform any thinking ability; instead, it performs a set of pre-defined functions.
• Artificial Super Intelligence (ASI): ASI can perform everything that a human can do. Alpha 2 is an example of ASI, which is the first humanoid ASI robot.

12. What Is Automatic Programming?

Automatic programming is describing what a program should do, and then having the AI system “write” the program.

13. Explain the different algorithms used for hyperparameter optimization.

Grid Search
Grid search trains the network for every combination by using the two set of hyperparameters, learning rate and the number of layers. Then evaluates the model by using Cross Validation techniques.

Random Search
It randomly samples the search space and evaluates sets from a particular probability distribution. For example, instead of checking all 10,000 samples, randomly selected 100 parameters can be checked.

Bayesian Optimization
This includes fine-tuning the hyperparameters by enabling automated model tuning. The model used for approximating the objective function is called the surrogate model (Gaussian Process). Bayesian Optimization uses Gaussian Process (GP) function to get posterior functions to make predictions based on prior functions.

14. What are Bayesian networks?

A Bayesian network is a probabilistic graphical model based on a set of variables and their dependencies, represented in the form of an acyclic graph. Bayesian networks are based on probability distribution, and they predict outcomes and detect anomalies using probability theory. The Bayesian networks are used to perform tasks such as prediction, detecting anomalies, reasoning, gaining insights, diagnostics, and decision-making. A Bayesian network, for example, could be used to illustrate the probability correlations between diseases and symptoms. The network may be used to calculate the chances of certain diseases being present based on symptoms.

15. What are the techniques used to avoid overfitting?

If we can detect overfitting at an early stage, it will be very useful for our training model. There are several methods up our sleeves that can be used to avoid overfitting-

• Cross-validation: Cross-validation is a resampling technique for evaluating machine learning models on a small sample of data.
• Remove features: We can remove the unnecessary features of the models to encompass the outliers.
• Early stopping: Early stopping is a type of regularization used in machine learning to minimize overfitting when using an iterative method like gradient descent to train a learner. Early stopping criteria specify how many iterations can be completed before the learner becomes over-fit.
• Training with more data: We can train our model with more data to accommodate outliers.
• Regularization: In machine learning, regularization is a method to solve the over-fitting problem by adding a penalty term with the cost function.
• Ensembling: Ensemble learning refers to combining the predictions from two or more models.