CAIC 9.2 Flashcards
What is the primary goal of retailers’ marketing campaigns?
To attract potential customers with incentives or discounts based on demographics.
What do ML models optimize in marketing campaigns?
The effectiveness of marketing campaigns by targeting the right customers.
What is the purpose of customer segmentation in marketing?
To understand different customer segments and improve marketing campaign effectiveness.
How do highly personalized marketing campaigns work?
By creating accurate individual profiles using large amounts of individual behavior data.
Fill in the blank: Contextual advertising is a targeted marketing technique that displays ads relevant to the _______.
content on a web page.
What is the role of generative AI in retail marketing?
To create dynamically personalized content tailored to individual customers’ preferences.
Why is understanding consumer perception crucial for retail businesses?
It significantly impacts their success and helps monitor brand reputation.
What is sentiment analysis?
A text classification problem that determines whether sentiment is positive, negative, or neutral.
How do ML algorithms assist in sentiment analysis?
They can classify new text data, such as social media posts, to understand overall sentiment.
What traditional methods are used for demand forecasting?
Buyer surveys, expert opinions, and projections based on past demands.
What statistical techniques are retailers using to improve demand forecasting?
Regression analysis and deep learning.
What is the significance of deep learning in demand forecasting?
It can incorporate multiple data sources to create more accurate demand forecasts.
What are point forecasts and probabilistic forecasts in ML?
Point forecasts provide a specific number, while probabilistic forecasts include a confidence score.
What role do AI and ML play in the automotive industry?
They improve efficiency, safety, and customer experience.
What are the three main stages of the system architecture for autonomous vehicles?
1) Perception and localization 2) Decision and planning 3) Control.
What is the function of the perception stage in autonomous driving?
To gather information about surroundings and determine the vehicle’s position.
What types of sensors are used in the perception stage of autonomous vehicles?
RADAR, LIDAR, cameras, and ultrasonic systems.
Fill in the blank: The decision and planning stage in autonomous driving acts as the _______ of the vehicle.
brain.
What is one application of AI/ML in the control module of autonomous vehicles?
Adaptive control systems that adjust control inputs based on sensor data.
How do reinforcement learning techniques benefit autonomous vehicles?
They enable vehicles to learn optimal control policies through trial and error.
What does ADAS stand for?
Advanced Driver Assistance Systems.
What are some functions of ADAS?
Detecting hazards, issuing warnings, and taking corrective actions.
What is the role of ML solutions architects in relation to ML algorithms?
To understand common real-world ML algorithms and design technology infrastructure.
What is an objective function in ML?
A business metric that the algorithm aims to minimize or maximize.
What is optimization in the context of ML algorithms?
The process of adjusting model parameters to minimize the disparity between projected and actual values.
Fill in the blank: The loss function used in optimization is often referred to as the _______.
objective function.
What is gradient descent?
An iterative approach for optimizing ML algorithms by calculating error change.
What does the learning rate control in ML optimization?
The magnitude of parameter updates at each iteration.
What is gradient descent?
An iterative approach for optimizing neural networks and ML algorithms by calculating the rate of error change.
What are the key parameters updated in gradient descent?
W and B (model parameters).
What controls the magnitude of parameter updates in gradient descent?
The learning rate.
List the key steps in the gradient descent optimization process.
- Initialize W randomly
- Calculate error using W
- Compute gradient of error with respect to loss function
- Update W to reduce error
- Repeat until gradient is zero.
What is the normal equation in the context of ML algorithms?
A one-step analytical solution for calculating coefficients of linear regression models.
What are the main types of ML problems discussed?
- Classification
- Regression.
What is classification in ML?
A task that assigns categories or classes to data points.
What is regression in ML?
A technique used to predict continuous numeric values.
What are important factors to consider when selecting an ML algorithm?
- Problem type
- Dataset size
- Number and nature of features
- Computational requirements
- Interpretability
- Assumptions about data distribution.
What is overfitting in machine learning?
When a trained model learns the training data too well but fails to generalize to new data.
What does linear regression aim to estimate?
The output value by calculating the weighted sum of input variables.
What is the formula for linear regression?
Y = W * X + B.
What is logistic regression used for?
Estimating the probability of an event occurring, such as transaction fraud.
How does logistic regression differ from linear regression?
Logistic regression uses a logistic function to map input variables to a probability score.
What is a decision tree?
A model that divides data hierarchically based on rules to make predictions.
What are the algorithms used for splitting in decision trees?
- Gini purity index
- Information gain.
What is a major advantage of decision trees?
Their ability to capture non-linear relationships and interactions between features.
What is a limitation of decision trees?
They can be sensitive to outliers and prone to overfitting.
What is random forest in machine learning?
An ensemble method that combines multiple decision trees to improve overall performance.
How does random forest reduce overfitting?
By introducing randomness in the model and using diverse subsets of features.
What is the key difference between random forests and gradient boosting?
Random forests use parallel independent weak learners, while gradient boosting employs a sequential approach.
What is a key advantage of gradient boosting?
It excels in handling imbalanced datasets and can achieve higher performance when properly tuned.
What is one of the limitations of gradient boosting?
It lacks parallelization capabilities, making it slower in training.
What is gradient boosting?
A machine learning technique that builds models sequentially to improve performance through tuning.
What are the advantages of gradient boosting?
- Higher performance when tuned properly
- Supports custom loss functions
- Captures complex relationships in data
- Produces accurate predictions
What are the limitations of gradient boosting?
- Lacks parallelization capabilities
- Sensitive to noisy data and outliers
- Less interpretable compared to simpler algorithms
What is XGBoost?
A popular implementation of gradient boosting known for faster training times and improved performance.
What improvements does XGBoost offer over standard gradient boosting?
- Training across multiple cores and CPUs
- Powerful regularization techniques
- Handles sparse datasets effectively
What are other popular variations of gradient boosting trees?
- LightGBM
- CatBoost
What is K-NN used for?
Both classification and regression tasks.
What is the underlying assumption of K-NN?
Similar items tend to have close proximity in the feature space.
How does K-NN classify a new data point?
By calculating distances to existing data points and using majority voting among the K nearest neighbors.
What is the impact of the choice of K in K-NN?
Significantly affects the performance of the model.
What are the limitations of K-NN?
- Complexity grows with data points
- Not suitable for high-dimensional datasets
- Sensitive to noisy and missing data
What is an artificial neuron?
A computational unit that processes inputs and transforms them using an activation function.
What is the function of the activation function in an artificial neuron?
Modifies the output of the linear function to capture non-linear relationships.
What does MLP stand for?
Multi-Layer Perceptron.
What is the architecture of an MLP?
Consists of an input layer, hidden layers, and an output layer.
What is the purpose of backpropagation in MLP training?
To adjust the weights of neurons based on their contribution to the total error.
What are the strengths of MLP?
- Suitable for classification and regression
- Captures intricate nonlinear patterns
- Efficient computational processing
What is clustering?
A data mining method that groups items based on shared attributes.
What is K-means clustering?
An unsupervised algorithm that groups data points into K clusters based on similarity.
What are the steps involved in K-means clustering?
- Randomly assign K centroids
- Adjust data point assignments to nearest centroids
- Update centroids to mean of assigned data points
- Repeat until convergence
What are the advantages of K-means clustering?
- Simplicity and ease of understanding
- Computationally efficient
- Interpretable clusters
What are the limitations of K-means clustering?
- Subjective selection of optimal K
- Sensitive to initial centroid placement
- Assumes spherical clusters with equal variance
- Sensitive to outliers
What is a time series?
A sequence of data points recorded at successive time intervals.
What is the purpose of time series analysis?
To analyze past patterns and predict future trends.
