#### Explain the terms Artificial Intelligence (AI), Machine Learning (ML and Deep Learning?

Artificial Intelligence (AI) is the domain of producing intelligent machines. ML refers to systems that can assimilate from experience (training data) and Deep Learning (DL) states to systems that learn from experience on large data sets. ML can be considered as a subset of AI. Deep Learning (DL) is ML but useful to large data sets. The figure below roughly encapsulates the relation between AI, ML, and DL:

In summary, DL is a subset of ML & both were the subsets of AI.

Additional Information: ASR (Automatic Speech Recognition) & NLP (Natural Language Processing) fall under AI and overlay with ML & DL as ML is often utilized for NLP and ASR tasks.

#### What are the different types of Learning/ Training models in ML?

ML algorithms can be primarily classified depending on the presence/absence of target variables.

A. Supervised learning: [Target is present]

The machine learns using labelled data. The model is trained on an existing data set before it starts making decisions with the new data.

The target variable is continuous: Linear Regression, polynomial Regression, quadratic Regression.

The target variable is categorical: Logistic regression, Naive Bayes, KNN, SVM, Decision Tree, Gradient Boosting, ADA boosting, Bagging, Random forest etc.

B. Unsupervised learning: [Target is absent]

The machine is trained on unlabelled data and without any proper guidance. It automatically infers patterns and relationships in the data by creating clusters. The model learns through observations and deduced structures in the data.

Principal component Analysis, Factor analysis, Singular Value Decomposition etc.

C. Reinforcement Learning:

The model learns through a trial and error method. This kind of learning involves an agent that will interact with the environment to create actions and then discover errors or rewards of that action.

#### What is the difference between deep learning and machine learning?

Machine Learning involves algorithms that learn from patterns of data and then apply it to decision making. Deep Learning, on the other hand, is able to learn through processing data on its own and is quite similar to the human brain where it identifies something, analyze it, and makes a decision.

The key differences are as follow:

- The manner in which data is presented to the system.
- Machine learning algorithms always require structured data and deep learning networks rely on layers of artificial neural networks.

Machine Learning Interview Question

#### What is the main key difference between supervised and unsupervised machine learning?

Supervised learning technique needs labeled data to train the model. For example, to solve a classification problem (a supervised learning task), you need to have label data to train the model and to classify the data into your labeled groups. Unsupervised learning does not need any labelled dataset. This is the main key difference between supervised learning and unsupervised learning.

#### How do you select important variables while working on a data set?

There are various means to select important variables from a data set that include the following:

- Identify and discard correlated variables before finalizing on important variables
- The variables could be selected based on ‘p’ values from Linear Regression
- Forward, Backward, and Stepwise selection
- Lasso Regression
- Random Forest and plot variable chart
- Top features can be selected based on information gain for the available set of features.

#### There are many machine learning algorithms till now. If given a data set, how can one determine which algorithm to be used for that?

Machine Learning algorithm to be used purely depends on the type of data in a given dataset. If data is linear then, we use linear regression. If data shows non-linearity then, the bagging algorithm would do better. If the data is to be analyzed/interpreted for some business purposes then we can use decision trees or SVM. If the dataset consists of images, videos, audios then, neural networks would be helpful to get the solution accurately.

So, there is no certain metric to decide which algorithm to be used for a given situation or a data set. We need to explore the data using EDA (Exploratory Data Analysis) and understand the purpose of using the dataset to come up with the best fit algorithm. So, it is important to study all the algorithms in detail.

Advance Machine Learning Interview Question

#### How are covariance and correlation different from one another?

Covariance measures how two variables are related to each other and how one would vary with respect to changes in the other variable. If the value is positive it means there is a direct relationship between the variables and one would increase or decrease with an increase or decrease in the base variable respectively, given that all other conditions remain constant.

Correlation quantifies the relationship between two random variables and has only three specific values, i.e., 1, 0, and -1.

1 denotes a positive relationship, -1 denotes a negative relationship, and 0 denotes that the two variables are independent of each other.

#### State the differences between causality and correlation?

Causality applies to situations where one action, say X, causes an outcome, say Y, whereas Correlation is just relating one action (X) to another action(Y) but X does not necessarily cause Y.

#### We look at machine learning software almost all the time. How do we apply Machine Learning to Hardware?

We have to build ML algorithms in System Verilog which is a Hardware development Language and then program it onto an FPGA to apply Machine Learning to hardware.

Machine Learning Interview Question

#### Explain One-hot encoding and Label Encoding. How do they affect the dimensionality of the given dataset?

One-hot encoding is the representation of categorical variables as binary vectors. Label Encoding is converting labels/words into numeric form. Using one-hot encoding increases the dimensionality of the data set. Label encoding doesn’t affect the dimensionality of the data set. One-hot encoding creates a new variable for each level in the variable whereas, in Label encoding, the levels of a variable get encoded as 1 and 0.

#### When does regularization come into play in Machine Learning?

At times when the model begins to underfit or overfit, regularization becomes necessary. It is a regression that diverts or regularizes the coefficient estimates towards zero. It reduces flexibility and discourages learning in a model to avoid the risk of overfitting. The model complexity is reduced and it becomes better at predicting.

#### What is Bias, Variance and what do you mean by Bias-Variance Tradeoff?

Both are errors in Machine Learning Algorithms. When the algorithm has limited flexibility to deduce the correct observation from the dataset, it results in bias. On the other hand, variance occurs when the model is extremely sensitive to small fluctuations.

If one adds more features while building a model, it will add more complexity and we will lose bias but gain some variance. In order to maintain the optimal amount of error, we perform a tradeoff between bias and variance based on the needs of a business.

Bias stands for the error because of the erroneous or overly simplistic assumptions in the learning algorithm . This assumption can lead to the model underfitting the data, making it hard for it to have high predictive accuracy and for you to generalize your knowledge from the training set to the test set.

Variance is also an error because of too much complexity in the learning algorithm. This can be the reason for the algorithm being highly sensitive to high degrees of variation in training data, which can lead your model to overfit the data. Carrying too much noise from the training data for your model to be very useful for your test data.

The bias-variance decomposition essentially decomposes the learning error from any algorithm by adding the bias, the variance and a bit of irreducible error due to noise in the underlying dataset. Essentially, if you make the model more complex and add more variables, you’ll lose bias but gain some variance — in order to get the optimally reduced amount of error, you’ll have to trade off bias and variance. You don’t want either high bias or high variance in your model.

Advance Machine Learning Interview Question

#### How can we relate standard deviation and variance?

Standard deviation refers to the spread of your data from the mean. Variance is the average degree to which each point differs from the mean i.e. the average of all data points. We can relate Standard deviation and Variance because it is the square root of Variance.

#### A data set is given to you and it has missing values which spread along 1standard deviation from the mean. How much of the data would remain untouched?

It is given that the data is spread across mean that is the data is spread across an average. So, we can presume that it is a normal distribution. In a normal distribution, about 68% of data lies in 1 standard deviation from averages like mean, mode or median. That means about 32% of the data remains uninfluenced by missing values.

#### Is a high variance in data good or bad?

Higher variance directly means that the data spread is big and the feature has a variety of data. Usually, high variance in a feature is seen as not so good quality.

Machine Learning Interview Question

#### If your dataset is suffering from high variance, how would you handle it?

For datasets with high variance, we could use the bagging algorithm to handle it. Bagging algorithm splits the data into subgroups with sampling replicated from random data. After the data is split, random data is used to create rules using a training algorithm. Then we use polling technique to combine all the predicted outcomes of the model.

#### A data set is given to you about utilities fraud detection. You have built a classifier model and achieved a performance score of 98.5%. Is this a good model? If yes, justify. If not, what can you do about it?

Data set about utilities fraud detection is not balanced enough i.e. imbalanced. In such a data set, accuracy score cannot be the measure of performance as it may only be predict the majority class label correctly but in this case our point of interest is to predict the minority label. But often minorities are treated as noise and ignored. So, there is a high probability of misclassification of the minority label as compared to the majority label. For evaluating the model performance in case of imbalanced data sets, we should use Sensitivity (True Positive rate) or Specificity (True Negative rate) to determine class label wise performance of the classification model. If the minority class label’s performance is not so good, we could do the following:

- We can use under sampling or over sampling to balance the data.
- We can change the prediction threshold value.
- We can assign weights to labels such that the minority class labels get larger weights.
- We could detect anomalies.

#### Explain the handling of missing or corrupted values in the given dataset.

An easy way to handle missing values or corrupted values is to drop the corresponding rows or columns. If there are too many rows or columns to drop then we consider replacing the missing or corrupted values with some new value.

Identifying missing values and dropping the rows or columns can be done by using Is Null() and dropna( ) functions in Pandas. Also, the Fillna() function in Pandas replaces the incorrect values with the placeholder value.

Advance Machine Learning Interview Question

#### What is Time series?

A Time series is a sequence of numerical data points in successive order. It tracks the movement of the chosen data points, over a specified period of time and records the data points at regular intervals. Time series doesn’t require any minimum or maximum time input. Analysts often use Time series to examine data according to their specific requirement.

#### What is a Box-Cox transformation?

Box-Cox transformation is a power transform which transforms non-normal dependent variables into normal variables as normality is the most common assumption made while using many statistical techniques. It has a lambda parameter which when set to 0 implies that this transform is equivalent to log-transform. It is used for variance stabilization and also to normalize the distribution.

#### What is the difference between stochastic gradient descent (SGD) and gradient descent (GD)?

Gradient Descent and Stochastic Gradient Descent are the algorithms that find the set of parameters that will minimize a loss function.

The difference is that in Gradient Descend, all training samples are evaluated for each set of parameters. While in Stochastic Gradient Descent only one training sample is evaluated for the set of parameters identified.

Machine Learning Interview Question

#### What is the exploding gradient problem while using back propagation technique?

When large error gradients accumulate and result in large changes in the neural network weights during training, it is called the exploding gradient problem. The values of weights can become so large as to overflow and result in NaN values. This makes the model unstable and the learning of the model to stall just like the vanishing gradient problem.

#### Can you mention some advantages and disadvantages of decision trees?

The advantages of decision trees are that they are easier to interpret, are nonparametric and hence robust to outliers, and have relatively few parameters to tune.

On the other hand, the disadvantage is that they are prone to overfitting.

#### Explain the differences between Random Forest and Gradient Boosting machines.

Random forests are a significant number of decision trees pooled using averages or majority rules at the end. Gradient boosting machines also combine decision trees but at the beginning of the process unlike Random forests. Random forest creates each tree independent of the others while gradient boosting develops one tree at a time. Gradient boosting yields better outcomes than random forests if parameters are carefully tuned but it’s not a good option if the data set contains a lot of outliers/anomalies/noise as it can result in overfitting of the model. Random forests perform well for multiclass object detection. Gradient Boosting performs well when there is data which is not balanced such as in real time risk assessment.

Advance Machine Learning Interview Question

#### What is a confusion matrix and why do you need it?

Confusion matrix (also called the error matrix) is a table that is frequently used to illustrate the performance of a classification model i.e. classifier on a set of test data for which the true values are well-known.

It allows us to visualize the performance of an algorithm/model. It allows us to easily identify the confusion between different classes. It is used as a performance measure of a model/algorithm.

A confusion matrix is known as a summary of predictions on a classification model. The number of right and wrong predictions were summarized with count values and broken down by each class label. It gives us information about the errors made through the classifier and also the types of errors made by a classifier.

#### What’s a Fourier transform?

Fourier Transform is a mathematical technique that transforms any function of time to a function of frequency. Fourier transform is closely related to Fourier series. It takes any time-based pattern for input and calculates the overall cycle offset, rotation speed and strength for all possible cycles. Fourier transform is best applied to waveforms since it has functions of time and space. Once a Fourier transform applied on a waveform, it gets decomposed into a sinusoid.

#### What do you mean by Associative Rule Mining (ARM)?

Associative Rule Mining is one of the techniques to discover patterns in data like features (dimensions) which occur together and features (dimensions) which are correlated. It is mostly used in Market-based Analysis to find how frequently an itemset occurs in a transaction. Association rules have to satisfy minimum support and minimum confidence at the very same time. Association rule generation generally comprised of two different steps:

- “A min support threshold is given to obtain all frequent item-sets in a database.”
- “A min confidence constraint is given to these frequent item-sets in order to form the association rules.”

Support is a measure of how often the “item set” appears in the data set and Confidence is a measure of how often a particular rule has been found to be true.

Machine Learning Interview Question

#### What is Marginalization? Explain the process.

Marginalization is summing the probability of a random variable X given joint probability distribution of X with other variables. It is an application of the law of total probability.

P(X=x) = ∑_{Y}P(X=x,Y)

Given the joint probability P(X=x,Y), we can use marginalization to find P(X=x). So, it is to find distribution of one random variable by exhausting cases on other random variables.

#### Explain the phrase “Curse of Dimensionality”.

The Curse of Dimensionality refers to the situation when your data has too many features.

The phrase is used to express the difficulty of using brute force or grid search to optimize a function with too many inputs.

It can also refer to several other issues like:

- If we have more features than observations, we have a risk of overfitting the model.
- When we have too many features, observations become harder to cluster. Too many dimensions cause every observation in the dataset to appear equidistant from all others and no meaningful clusters can be formed.

Dimensionality reduction techniques like PCA come to the rescue in such cases.

#### What is the Principle Component Analysis?

The idea here is to reduce the dimensionality of the data set by reducing the number of variables that are correlated with each other. Although the variation needs to be retained to the maximum extent.

The variables are transformed into a new set of variables that are known as Principal Components’. These PCs are the eigenvectors of a covariance matrix and therefore are orthogonal.

Advance Machine Learning Interview Question

#### Why is rotation of components so important in Principle Component Analysis (PCA)?

Rotation in PCA is very important as it maximizes the separation within the variance obtained by all the components because of which interpretation of components would become easier. If the components are not rotated, then we need extended components to describe variance of the components.

#### What are outliers? Mention three methods to deal with outliers.

A data point that is considerably distant from the other similar data points is known as an outlier. They may occur due to experimental errors or variability in measurement. They are problematic and can mislead a training process, which eventually results in longer training time, inaccurate models, and poor results.

The three methods to deal with outliers are:

Univariate method – looks for data points having extreme values on a single variable

Multivariate method – looks for unusual combinations on all the variables

Minkowski error – reduces the contribution of potential outliers in the training process

#### What is the difference between regularization and normalization?

Normalization adjusts the data; regularization adjusts the prediction function. If your data is on very different scales (especially low to high), you would want to normalize the data. Alter each column to have compatible basic statistics. This can be helpful to make sure there is no loss of accuracy. One of the goals of model training is to identify the signal and ignore the noise if the model is given free rein to minimize error, there is a possibility of suffering from overfitting. Regularization imposes some control on this by providing simpler fitting functions over complex ones.

Machine Learning Interview Question

#### Explain the difference between Normalization and Standardization.

Normalization and Standardization are the two very popular methods used for feature scaling. Normalization refers to re-scaling the values to fit into a range of [0,1]. Standardization refers to re-scaling data to have a mean of 0 and a standard deviation of 1 (Unit variance). Normalization is useful when all parameters need to have the identical positive scale however the outliers from the data set are lost. Hence, standardization is recommended for most applications.

#### List the most popular distribution curves along with scenarios where you will use them in an algorithm.

The most popular distribution curves are as follows- Bernoulli Distribution, Uniform Distribution, Binomial Distribution, Normal Distribution, Poisson Distribution, and Exponential Distribution.

Each of these distribution curves is used in various scenarios.

Bernoulli Distribution can be used to check if a team will win a championship or not, a newborn child is either male or female, you either pass an exam or not, etc.

Uniform distribution is a probability distribution that has a constant probability. Rolling a single dice is one example because it has a fixed number of outcomes.

Binomial distribution is a probability with only two possible outcomes, the prefix ‘bi’ means two or twice. An example of this would be a coin toss. The outcome will either be heads or tails.

Normal distribution describes how the values of a variable are distributed. It is typically a symmetric distribution where most of the observations cluster around the central peak. The values further away from the mean taper off equally in both directions. An example would be the height of students in a classroom.

Poisson distribution helps predict the probability of certain events happening when you know how often that event has occurred. It can be used by businessmen to make forecasts about the number of customers on certain days and allows them to adjust supply according to the demand.

Exponential distribution is concerned with the amount of time until a specific event occurs. For example, how long a car battery would last, in months.

#### How do we check the normality of a data set or a feature?

Visually, we can check it using plots. There is a list of Normality checks, they are as follow:

- Shapiro-Wilk W Test
- Anderson-Darling Test
- Martinez-Iglewicz Test
- Kolmogorov-Smirnov Test
- D’Agostino Skewness Test

Advance Machine Learning Interview Question

#### What is Linear Regression?

Linear Function can be defined as a Mathematical function on a 2D plane as, Y =Mx +C, where Y is a dependent variable and X is Independent Variable, C is Intercept and M is slope and same can be expressed as Y is a Function of X or Y = F(x).

At any given value of X, one can compute the value of Y, using the equation of Line. This relation between Y and X, with a degree of the polynomial as 1 is called Linear Regression.

In Predictive Modeling, LR is represented as Y = Bo + B1x1 + B2x2

The value of B1 and B2 determines the strength of the correlation between features and the dependent variable.

Example: Stock Value in $ = Intercept + (+/-B1)*(Opening value of Stock) + (+/-B2)*(Previous Day Highest value of Stock)

#### Differentiate between regression and classification.

Regression and classification are categorized under the same umbrella of supervised machine learning. The main difference between them is that the output variable in the regression is numerical (or continuous) while that for classification is categorical (or discrete).

Example: To predict the definite Temperature of a place is Regression problem whereas predicting whether the day will be Sunny cloudy or there will be rain is a case of classification.

#### What is target imbalance? How do we fix it? A scenario where you have performed target imbalance on data. Which metrics and algorithms do you find suitable to input this data onto?

If you have categorical variables as the target when you cluster them together or perform a frequency count on them if there are certain categories which are more in number as compared to others by a very significant number. This is known as the target imbalance.

Example: Target column – 0,0,0,1,0,2,0,0,1,1 [0s: 60%, 1: 30%, 2:10%] 0 are in majority. To fix this, we can perform up-sampling or down-sampling. Before fixing this problem let’s assume that the performance metrics used was confusion metrics. After fixing this problem we can shift the metric system to AUC: ROC. Since we added/deleted data [up sampling or downsampling], we can go ahead with a stricter algorithm like SVM, Gradient boosting or ADA boosting.

Machine Learning Interview Question

#### List all assumptions for data to be met before starting with linear regression.

Before starting linear regression, the assumptions to be met are as follow:

- Linear relationship
- Multivariate normality
- No or little multicollinearity
- No auto-correlation
- Homoscedasticity

#### When does the linear regression line stop rotating or finds an optimal spot where it is fitted on data?

A place where the highest RSquared value is found, is the place where the line comes to rest. RSquared represents the amount of variance captured by the virtual linear regression line with respect to the total variance captured by the dataset.

#### Why is logistic regression a type of classification technique and not a regression? Name the function it is derived from?

Since the target column is categorical, it uses linear regression to create an odd function that is wrapped with a log function to use regression as a classifier. Hence, it is a type of classification technique and not a regression. It is derived from cost function.

Advance Machine Learning Interview Question

#### What could be the issue when the beta value for a certain variable varies way too much in each subset when regression is run on different subsets of the given dataset?

Variations in the beta values in every subset implies that the dataset is heterogeneous. To overcome this problem, we can use a different model for each of the clustered subsets of the dataset or use a non-parametric model such as decision trees.

#### What does the term Variance Inflation Factor mean?

Variation Inflation Factor (VIF) is the ratio of variance of the model to variance of the model with only one independent variable. VIF gives the estimate of volume of multicollinearity in a set of many regression variables.

VIF = Variance of model Variance of model with one independent variable

#### Which machine learning algorithm is known as the lazy learner and why is it called so?

KNN is a Machine Learning algorithm known as a lazy learner. K-NN is a lazy learner because it doesn’t learn any machine learnt values or variables from the training data but dynamically calculates distance every time it wants to classify, hence memorises the training dataset instead.

Machine Learning Interview Question

#### Is it possible to use KNN for image processing?

Yes, it is possible to use KNN for image processing. It can be done by converting the 3-dimensional image into a single-dimensional vector and using the same as input to KNN.

#### Differentiate between K-Means and KNN algorithms?

KNN is Supervised Learning where-as K-Means is Unsupervised Learning. With KNN, we predict the label of the unidentified element based on its nearest neighbor and further extend this approach for solving classification/regression-based problems.

K-Means is Unsupervised Learning, where we don’t have any Labels present, in other words, no Target Variables and thus we try to cluster the data based upon their coordinates and try to establish the nature of the cluster based on the elements filtered for that cluster.

#### How does the SVM algorithm deal with self-learning?

SVM has a learning rate and expansion rate which takes care of this. The learning rate compensates or penalises the hyperplanes for making all the wrong moves and expansion rate deals with finding the maximum separation area between classes.

Advance Machine Learning Interview Question

#### What are Kernels in SVM? List popular kernels used in SVM along with a scenario of their applications.

The function of kernel is to take data as input and transform it into the required form. A few popular Kernels used in SVM are as follows: RBF, Linear, Sigmoid, Polynomial, Hyperbolic, Laplace, etc.

#### What is Kernel Trick in an SVM Algorithm?

Kernel Trick is a mathematical function which when applied on data points, can find the region of classification between two different classes. Based on the choice of function, be it linear or radial, which purely depends upon the distribution of data, one can build a classifier.