CAIC 1 9 part 1 Flashcards
(26 cards)
How do personalized marketing campaigns work?
They create accurate individual profiles using large amounts of individual behavior data, such as historical transaction data and social media data, to generate highly personalized campaigns for a higher conversion rate.
What role does AI and ML play in the automotive industry?
They are used to improve efficiency, safety, and overall transformation in the automotive industry.
What is the perception stage in autonomous driving?
It involves the autonomous vehicle (AV) gathering information about its surroundings using various sensors and determining its position relative to the environment.
What sensors are used in the perception stage of autonomous vehicles?
The AV employs sensors such as RADAR, LIDAR, cameras, and other systems to capture data from the surrounding environment.
What is a key component in the perception stage of autonomous vehicles?
The recognition module, which utilizes ML algorithms to detect and classify objects such as pedestrians and vehicles.
What considerations should be kept in mind when choosing a ML algorithm?
Consider the problem type, dataset size, and the number and nature of features in your dataset.
What are classification algorithms used for?
They are suitable for tasks where the goal is to categorize data into distinct classes.
What is overfitting in machine learning?
Overfitting occurs when a trained model learns the training data too well but fails to generalize to new, unseen data.
What is linear regression?
It is an algorithm developed to solve regression problems by predicting continuous values based on independent inputs.
What is the formula for linear regression?
The formula is expressed as Y = b + wx, where Y is the predicted value, b is the intercept, and wx is the linear combination of input variables.
What is logistic regression used for?
It is used for binary classification tasks and offers fast training speed and interpretability.
What is a limitation of decision trees?
They cannot extrapolate beyond the range of training inputs, which limits their predictive capability.
What is a random forest algorithm?
It is a machine learning algorithm that uses multiple decision trees to improve accuracy and reduce overfitting by combining predictions through majority voting or averaging.
What is an artificial neural network (ANN)?
An ANN consists of layers of interconnected neurons, including an input layer, hidden layers, and an output layer.
What is the purpose of a multi-layer perceptron (MLP)?
Each hidden layer learns higher-level representations of the previous layer’s features, capturing important information for predictions.
What is DeepAR?
It is a state-of-the-art forecasting algorithm based on neural networks, designed to handle large datasets with multiple similar target time series.
What is collaborative filtering?
It is a recommendation algorithm that suggests items to users based on the preferences of similar users.
What is matrix factorization in recommendation systems?
It involves learning vector representations for users and items to approximate the original user-item interaction matrix.
What is the vanishing gradient problem?
It occurs when signals from initial inputs diminish as they pass through multiple layers in a CNN, leading to performance issues.
How do ResNets address the vanishing gradient problem?
They implement skip connections that allow signals to bypass certain layers, preserving important information.
What is Word2Vec?
It is a technique for learning word embeddings, where CBOW predicts a word based on surrounding words, and continuous-skip-gram predicts surrounding words for a given word.
What is BERT?
BERT is a pre-trained model used for various NLP tasks, including text classification and question answering, and can be fine-tuned for specific tasks.
What is the difference between GANs and newer generative models?
Newer generative models can learn tasks with few or no examples, while GANs require extensive training data.
What are large language models (LLMs)?
LLMs are capable of generating text, translating languages, and providing informative answers, trained on extensive datasets.
