A convolutional neural network architecture designed for biomedical image segmentation.
Used for medical image segmentation in radiology.
A method in machine learning used to estimate the reliability of predictions.
Used in Bayesian neural networks for estimating confidence in predictions.
A statistical estimator whose expected value equals the true parameter value.
Sample mean as an unbiased estimator of the population mean.
A scenario in machine learning where a model is too simple and fails to capture underlying patterns in data.
A linear model failing to capture non-linear data relationships.
A technique in data preprocessing that reduces the majority class to balance a dataset.
Used in fraud detection to balance class distributions.
A machine learning paradigm where models learn patterns from unlabeled data.
Used in customer segmentation for marketing.
A probability distribution where all outcomes have equal probability.
Used in random weight initialization for neural networks.
A technique used to predict the incremental impact of an intervention or treatment.
Used in targeted marketing campaigns to measure ad effectiveness.
A mathematical theorem stating that neural networks can approximate any function given sufficient neurons and depth.
Used to justify the power of deep learning in complex tasks.
A recommendation system approach that suggests items based on user similarity.
Used in movie recommendation systems like Netflix.
A mathematical function used in decision-making models to measure the desirability of different outcomes.
Used in reinforcement learning to define reward functions.
Data that does not follow a predefined format, such as text, images, or videos.
Used in natural language processing and image recognition.
A reinforcement learning strategy used in multi-armed bandit problems to balance exploration and exploitation.
Used in online advertising for optimizing click-through rates.
A formula used in machine learning algorithms to adjust model parameters during training.
Used in neural networks to minimize loss functions.
Used in feature selection to reduce redundancy in datasets.
A statistical analysis technique that examines one variable at a time.
Used in exploratory data analysis to understand distributions.
A feature engineering metric that calculates the ratio of unique values to total observations.
Used in preprocessing to filter out categorical features with too many unique values.
Vector representations of user behavior used in recommendation systems.
Used in e-commerce platforms to recommend personalized products.
A probability distribution with a single peak.
Used in statistics to model natural phenomena like human height.
A dataset where one class has significantly more samples than another.
Common in fraud detection where fraudulent transactions are rare.
A technique in machine learning to quantify the confidence of a model’s predictions.
Used in autonomous driving to improve decision-making in uncertain environments.
A technique where a recurrent neural network (RNN) is expanded over multiple time steps for training.
Used in training Long Short-Term Memory (LSTM) networks for time-series data.
A probability distribution where all outcomes are equally likely.
Used in random initialization of weights in neural networks.
The property that a neural network can approximate any continuous function given enough neurons.
Used in deep learning for complex pattern recognition.
A problem in machine learning where a model is too simple to capture the underlying pattern in the data.
Occurs when using linear regression on non-linear data.
A technique used to increase the resolution or amount of data in a dataset.
Used in computer vision to enhance low-resolution images.
A type of machine learning where models find hidden patterns in unlabeled data.
Used in customer segmentation for marketing strategies.
A type of graph in which all edges have the same weight or no weight at all.
Used in social network analysis where relationships exist but have no strength value.
A sampling technique where every data point has an equal probability of being selected.
Used in data preprocessing to balance datasets.
A recommendation algorithm that suggests items to users based on the preferences of similar users.
Used in Netflix recommendations to suggest movies based on user preferences.
A function that measures the usefulness or reward of different outcomes in decision-making models.
Used in reinforcement learning to evaluate policy rewards.
A technique in active learning where a model queries the most uncertain data points for labeling.
Used in interactive AI systems to improve model training with limited labeled data.
A technique where iterative processes are expanded into explicit computational graphs for better optimization.
Used in meta-learning for optimizing gradient-based learning models.
A technique for scaling input features to have unit norm in machine learning models.
Used in text processing to normalize word vectors in NLP.
Data that does not have a predefined format or organization.
Used in deep learning models for sentiment analysis and image recognition.
A training method where models first learn from unlabeled data before fine-tuning on labeled data.
Used in NLP models like BERT and GPT for language understanding.
An exploration strategy in reinforcement learning that balances exploration and exploitation.
Used in multi-armed bandit problems to optimize online recommendations.
An estimator whose expected value is equal to the true parameter value in statistical inference.
Used in statistical machine learning for parameter estimation.
The analysis of a single variable without considering relationships with others.
Used in data preprocessing for understanding individual feature distributions.
A task in natural language processing (NLP) that categorizes user queries based on intent.
Used in virtual assistants to understand user requests.
ML is a subset of AI that enables machines to learn patterns from data and make predictions or decisions without explicit programming.
Spam detection in emails using classification models.
DL is a subset of ML that uses artificial neural networks to process complex data and perform high-level computations.
Image recognition in self-driving cars.
Gen AI refers to AI models that generate new content, including text, images, and code, using trained knowledge bases.
AI models like ChatGPT and Stable Diffusion that generate text and images.