Machine Learning Glossary

Noise

Noise in machine learning refers to irrelevant or meaningless data or distortions in a dataset.

Normalization

Normalization in machine learning is a preprocessing technique used to scale numerical data to a common range without distorting differences in the ranges of values.

Null Accuracy

Null accuracy is a baseline measure used in classification problems to evaluate the effectiveness of a model.

Object Detection

Object detection is a computer vision technique that involves identifying and locating objects within an image or a video.

Object Tracking

Object tracking in computer vision is the process of locating a moving object (or multiple objects) over time in a video sequence or in real-time applications.

Observation

In machine learning, an observation refers to a single instance of data in a dataset.

One-Hot Encoding

One-hot encoding is a process of converting categorical data into a form that could be provided to machine learning algorithms to improve predictions.

Optimizers

In machine learning, optimizers are algorithms or methods used to change the attributes of the neural network, such as weights and learning rate, to reduce the losses.

Outlier

An outlier in machine learning is a data point that differs significantly from other observations in the dataset.

Output Layer

The output layer in a neural network is the final layer that produces the result of the model.

Overfitting

Overfitting in machine learning occurs when a model is excessively complex and starts to learn both the underlying pattern and the noise in the training data.

Pandas and Numpy

Pandas and Numpy are fundamental Python libraries widely used in data science and machine learning.

Panoptic Segmentation

Panoptic segmentation is a computer vision task that combines elements of both semantic segmentation (identifying object classes in an image) and instance segmentation (distinguishing between different instances of the same class).

Parameter

In machine learning, a parameter is an internal variable of a model that is learned from the training data.

Parameter-Efficient Fine-Tuning

Parameter-efficient fine-tuning refers to techniques in machine learning that allow for effective fine-tuning of models, especially large ones like transformers, without needing to adjust a large number of parameters.

Part-of-Speech Tagging

Part-of-speech tagging is a process in natural language processing where parts of speech (like nouns, verbs, adjectives, etc.) are identified and assigned to each word in a given text.

Pascal

Pascal, in the context of machine learning, often refers to the PASCAL Visual Object Classes (VOC) challenge, a benchmark in computer vision that involves a series of datasets and challenges in object detection and image segmentation.

Pattern Matching

Pattern matching in computer science refers to the technique of checking a given sequence of tokens for the presence of the constituents of some pattern. It is a fundamental operation in many fields, including text processing and computer vision.

Pattern Recognition

Pattern recognition is a branch of machine learning that focuses on the recognition of patterns and regularities in data. It is used to classify data based on statistical information extracted from the patterns.

Polygon

In the context of computer vision and graphics, a polygon is a plane figure that is described by a finite number of straight line segments connected to form a closed polygonal chain or circuit.

Pooling Layers in CNN

Pooling layers in convolutional neural networks (CNNs) are used to reduce the spatial dimensions (width and height) of the input volume for the next convolutional layer.

Positive Class

In binary classification, the positive class is one of the two output classes, representing the presence or occurrence of a specific feature or characteristic.

Pre-trained Transformer

A pre-trained transformer is a type of neural network model that has been previously trained on a large dataset and can be fine-tuned for various specific tasks.

Precision

Precision in machine learning is a metric used to evaluate the accuracy of a classification model, defined as the ratio of true positive predictions to the total predicted positives (true positives plus false positives).

Predictive Model Validation

Predictive model validation is the process of confirming that a predictive model functions well with new, independent data.

Probabilistic Classification

Probabilistic classification in machine learning refers to classification models that predict a probability distribution over a set of classes, rather than just predicting the most likely class.