Make your own quiz Manage my Quizzes Plans / Features Guides and Help Contact Privacy & Security

IS4

A futuristic digital interface showcasing elements of artificial intelligence, including neural networks and data visualization concepts, with a sleek, modern aesthetic.

Deep Dive into Artificial Intelligence

Test your knowledge on the latest trends and principles in artificial intelligence with this comprehensive quiz. Covering various topics, from neural networks to data visualization, this quiz challenges your understanding of the core concepts.

Key Features:

53 thought-provoking questions
Multiple-choice format for easy answering
Engaging insights into AI technologies

53 Questions13 MinutesCreated by LearningMachine42

Artificial Intelligence can be applied to the following sectors

Robotics

Information Extraction

All the answer

Artificial neural networks are capable to learn human biases

False: the achievable complexity of the artificial neural networks is so far from the complexity of the human brain to make impossible to mimic this characteristic

False: human biases are not reproducible nor measurable

True

Nowadays, the usage of classical feature extraction and data analysis methods is outdated since the capability of the recent deep learning models and methods made them obsolete and not more present in the common practice.

True

False

Recent artificial intelligence models can solve analogy puzzles like “Paris is to France as Tokyo is to?” producing the correct answer “Japan.”

True

False

Considering the “Data knowledge spectrum plot” discussed in class, the minimum amount of data required is in the following case.

No knowledge about the model generating the data is available

A statistical model of the process is available

A mathematical model of the process is available

It is possible to think to the single datum in input to the neural network as a point in the “input space” of the model, even if the input is a single value, a N dimensional vector, or an image

True

False

It is correct to say the one of the key feature of an intelligent artificial system is the capability to learn (even if only a limited sense) and/or get better in time

True

False

According to the Andries Engelbrecht definition of Computational intelligence what of the following is not included?

Artificial Neural Networks

Evolutionary Computing

Swarm Intelligence

Artificial immune system

Fuzzy Systems

All the answers

According to the class discussion of the Gestalt capability, what of the following sentences is more correct?

The Gestalt capability is a typical feature present by-design in the model of classical neural networks

The Gestalt capability is a typical feature present by-design in the model of deep learning neural networks

The Gestalt capability is a typical human feature not well (yet) mimicked in current artificial networks

The following activity: a) Data Selection; b) Data Filtering; c) Data Enhancing…

1. Are part of the job of the artificial intelligent specialist in normal activities

Contribute to keep lower the complexity of the learning task

Normal activities & Lower Complexity

Are part of the classical machine learning approaches and they are (correctly) no longer used in deep learning applications

The Mean Squared Error is typically present in what step of the design?

Representation

Evaluation

Considering IoT devices as source of data for external intelligent systems (IS is not intended to be embedded into the IoT device), what kind of IoT devices can be really used?

Passive data IoT devices

Active data IoT devices

Dynamic data IoT devices

All the answers

None of the answers

Referring to the class discussion, the (correct) design practice for neural networks considers

Start with deep learning models since they are the cutting edge and most advanced technology we have now

Start with deep learning models since they are the cutting edge and most advanced technology we have now, and then use classicals method as reference

Start with simple neural networks before to consider deep learning models

The missing values can also be occupied by computing mean, mode or median of the observed given values.

This is very unusual and not common in practice

This is a very simple and effective solution in case the learning method is not capable to deal with missing data

This is not possible, since that is just descriptive statistics about the features, and cannot be used to fill missing data

Referring to the class discussion on data leakage what is the worst situation?

The unwanted leakage of data from test dataset to training data set

The unwanted leakage of data from training dataset to test data set

None of the above since transferring data from test and/or training dataset is normal when the accuracy of the model is tested

An additional information can allow the model to learn or know something that it otherwise would not know and in turn invalidate the estimated performance of the model being constructed. This is called

Data leakage

Data pre-processing

Data harmonization

Data wrangling

The degrees of freedom for a given problem are the number of independent problem variables which must be specified to uniquely determine a solution. Hence the #DoF is important to be considered

To design the number of vectors in the learning dataset.

To avoid overfitting problem in the model

All the answers

None of the answers

About the cosine metrics it is possible to say that

Two vectors with the same orientation have a cosine similarity of 1

Two vectors oriented at 90° relative to each other have a similarity of 0

All the answers

None of the answers

What similarity feature/features discussed in class offers/offer the property to allow a fast comparison based on a short 1D vector of elements or bits

Phash

Ahash

Both phash and ahash

Cross-Correlation

In agreement to the class discussion, which description better describes the design activity?

Similarity in the dataset requires more space and processing time

Similarity in the dataset can improve generalization

Both of the answers

None of the answers

In agreement to the class discussion, in a dataset of 1100 labelled images, the search for duplications is typically achieved…

By manual exploration of the dataset for better results since the number of images is not critical

By automatic iterations

In agreement to the class discussion, what kind of labelling error is generally the worst case for the accuracy of the generalization of the model? ERR1 = Duplications with same labels EER2 = Duplications with different labels

ERR1

ERR2

ERR1 = ERR2

According to the class discussion, about the relationship between the operation of cross-correlation and convolution it is possible to say that:

They are very similar in meaning and mathematical expression

Despite the mathematical expression is similar, the meaning and their use is completely different

There is no specific relationship since they are different in meaning and mathematical expressions

According to the class discussion, what is the characteristic of the self-correlation (O=xcor2(A,A)) map produced by a generic image?

A flat and noisy central plateau

An evident spike at the center with a very well defined maximum

It is not possible to create an autocorrelation map from one single images, two different images are needed

If your data set contains extreme outliers it better to use as preprocessing

Feature clipping

Min-max normalization

Z’ norm

A logarithmic scaling to one feature values is typically applied in a case of

Outliers presence

Negative values

A very large range in the values (>0)

According to the scientific visualization rules presented in class, if you are plotting many figures of merit obtained by your trained neural network on a new dataset, which is the correct ranking of visual attributes to be used? Left: low accuracy Right: HIGH ACCURACY

Color intensity > Hue > Length

Area > Length > Hue

Slope > Angle > Volume

Hue > Area > Length

According to the scientific visualization rules presented in class, is it possible to plot a graphical representation of the confidence level of your figures of merit of your trained model?

No, it is a statistical index with different units and meaning and hence can not be represented in the same plot

Yes, the confidence interval data have the same units and meaning and they can be represented in the same plot

According to the discussion presented in class about the data visualization, and considering the following steps of the design workflow 1) Get Data, 2) Clean Manipulate Data, 3) Train models, 4) Test Data, 5) Improve the design, which are the main step/steps where data visualization should be involved?

#1 and #5

#2, #3 and #5

According to the discussion presented in class about the similarity, consider an image A(x,y) with internal similarity (repetitions of patterns). What happens to the output of the self cross-correlation (O = xcorr2(A,A))

It is not possible to apply the cross-correlationto the same image

Output O tends to be a flat plateau with one clear central peak

Output O tends to have many peaks and one evident maximum

Output O tends to have many equivalent peaks with the same maximum value

You have a dataset X of 1000 samples and number of features F = 4 features with output labels. You want to reduce the number of features F to 2 for data visualization. According to the goal, consider the following options. OPTION A: Apply PCA to X, and select only the first 2 Principal Components OPTION B: Apply the Feedforward Feature Selection to X, and select only the first 2 more relevant features

Option A is possible. Option B is possible

Option A is NOT possible. Option B is possible

Option A is possible. Option B is NOT possible

Option A is NOT possible. Option B is NOT possible

You have a feature in your dataset with the following values F1 = [ -5 0 +5], which normalization will give you the following F1_norm = [0 0.5 1]

Min-MAX

Z-score

Clipping

A different type of normalization

According to the class discussion, in general for a given small dataset X with output labels, if you train a feed-forward neural models (of the same type) with an increasing number of neurons, which case is more probable?

None of the answers

The training error and the validation will decrease indefinitely

The training error will increase

The validation error will decrease indefinitely

According to the class discussion, in a cross-validation single test, which train/test partition of the samples will provide the lower training error but the lower confidence in the test results?

Training set = 99% , Test Set = 01%

2. Training set = 75% , Test Set = 25%

Training set = 50% , Test Set = 50%

Training set = 25% , Test Set = 75%

Training set = 01% , Test Set = 99%

According to the class discussion, what kind of activity can be performed on the test set?

All of the answers

Mean test error estimation

Mean test error estimation and standard deviation

Confusion matrix test

According to the class discussion, what kind of activity can be performed on the train set?

All the other options

Design of the #of neurons

Design of the #of layers

Normalization

PCA

According to the class discussion, where can be performed the feature engineering?

Only on the train set

Only on the test set

On the train set and the test set

Not on the train, Not on the test set, but only on a different dataset

A simple k-Fold Cross Validation procedure may

Lead to disarranging the proportion of examples from each class in the test partitions

Making impossible to process the test error

Get stuck into one the local minima

Produce severe overfitting

None of the answers

Which option is correct?

From the confusion matrix is possible to process the classification error

From the confusion matrix is possible to process the classification error and vice versa

The confusion matrix is applicable only to binary classification systems

The classification error is equal to the sum of the diagonal elements of the confusion matrix

According to the notation used in class, which kind of a model is described by the equation f(x) = sgn(w * x + b)

Linear classifier

Linear regressor

Soft-max neuron

Sigmoidal neuron

Gradient descent formula

Number of the model's parameters

According to the notation used in class, which kind of a classifier is better described by the following definition: “the output is the label produced by the most probable classifier”

Bayes Optimal Classifier

Supervised Classifier

K-means

None of the answers

According to the class discussion the kNN classifier, what kind of learning is it?

Instance-based Learning

Eager Learning

Hard-limited Learning

Unsupervised Clustering

None of the answers

According to the class discussion, what is the classifier with the following properties: 1) not based on neural techniques; 2) It’s deterministic with no random initialization; 3) Perfect repeatability; 4) A minimum number of parameters is needed; 5) Learning is very simple but effective; 6) Perfect explainability

KNN

Linear classifier

Decision Tree

K-means

None of the answers

According to the class discussion on kNN classifiers about the k parameter and its relationship to regularization of the decision boundaries and the computational complexity, what is the correct option about larger values of k?

More regulatization and more complexity

Less regulatization and more complexity

More regulatization and less complexity

Less regulatization and less complexity

The parameter k is not related to regularization and complexity

According to the class discussion on PCA what is the correct option?

PCA vectors are originating from the center of mass of the points

All subsequent principal component vectors are orthogonal

All the other options

According to the class discussion on PCA what is the correct option?

All subsequent principal component vectors are orthogonal

The variance of the data projection on the first PCA vectors is maximized

All the other options

According to the class discussion about unsupervised learning , what is the method with the following properties: • You need to specify the number of clusters k in advance • Is unable to handle noisy data and outliers • It is not suitable to discover clusters with non-convex shapes

K-means

KNN

Decision Tree

None of the answers

According to the class discussion, considering the equation of the back-propagation in a feedforward neural network of weight w_ijconnected to the following output neuron k, which is the missing term? DELTAW_ij= ??? * y_j* delta_k

??? = alfa (the regularization term < 1)

??? = alfa (the regularization term > 1)

??? = x_j(the input vector)

??? = x_j(the input vector error)

According to the class discussion, considering a standard intelligent vision system, which capability can be processed onboard on a recent smart industrial camera?

Segmentation

Segmentation, Measurement

Segmentation, Measurement, Classification with trained non-deep models

Segmentation, Measurement, Classification with trained deep models

Segmentation, Measurement, Classification with trained deep models and training of deep models

According to the class discussion, Traditional Segmentation methods are quite useful to produce blobs or object candidates to be further processed by deep models for classification or measurements. Traditional Segmentation methods can be partitioned in

Global knowledge, Edge-based

Edge-based, Region-based

Global knowledge, Edge-based, Region-based

None of the answers

According to the class discussion, considering a general CNN architecture, what is the sequence of modules which is more likely

Input layer  Convolution  Relu  Max Pooling …  Softmax  Output

Input layer  Relu  Convolution  Max Pooling …  Softmax  Output layer

Input layer  Relu  Max Pooling Convolution …  Softmax  Output layer

Input layer  Relu  Max Pooling  Softmax  Convolution …  Output layer

According to the class discussion referred to edge computing, is it possible to process images with trained deep learning models on external small dedicated devices connect via USB connection?

True: the usage of dedicated processors and the USB bandwidth make this option

False: the USB bandwidth make this option not possible

False: the needed computational complexity needed to run trained deep learning models make this option not possible

False: the bandwidth and the computational complexity need to process images with trained deep learning model is not adequate

According to the class discussion what is Greedy Layer-Wise Training?

A supervised training step to improve auto-encoders

A supervised training step to classical feedforward networks

An unsupervised training step to classical feedforward networks

An unsupervised training step to improve auto-encoders

{"name":"IS4", "url":"https://www.quiz-maker.com/QPREVIEW","txt":"Test your knowledge on the latest trends and principles in artificial intelligence with this comprehensive quiz. Covering various topics, from neural networks to data visualization, this quiz challenges your understanding of the core concepts. Key Features: 53 thought-provoking questions Multiple-choice format for easy answering Engaging insights into AI technologies","img":"https:/images/course3.png"}