Policy gradient theorem - The policy gradient theorem expresses how an agent's expected cumulative reward changes when the parameters of its action-selection policy change.
Policy improvement - Policy improvement is the process of changing a decision policy using information about the value of its actions.
Policy iteration - Policy iteration is a method for finding an improved decision policy by repeatedly alternating policy evaluation and policy improvement.
Policy loss - Policy loss is the shortfall in expected performance caused by following a particular policy instead of an optimal policy for the same decision problem.
Policy search - Policy search is a reinforcement-learning approach that improves a decision rule by searching over candidate policies or their parameters.
Policy value - Policy value is the expected return obtained by following a specified policy under a specified decision problem.
Poly-Weibull distribution - The poly-Weibull distribution is a continuous time-to-event distribution constructed by adding several Weibull-type hazard functions.
Pólya urn model - A Pólya urn model is a probability model in which sampling changes the probabilities of future samples through a specified reinforcement rule.
Polyak-Ruppert averaging - Polyak-Ruppert averaging is a stochastic-approximation technique that averages successive parameter estimates rather than returning only the last estimate.
Polyglot persistence - Polyglot persistence is the use of different data-storage technologies within an application or organization to suit different data and access needs.
Polynomial fitting - Polynomial fitting estimates a relationship using a sum of powers of an input variable, such as a + bx + cx squared.
Polynomial kernel - A polynomial kernel compares two feature vectors by raising a scaled, shifted dot product to an integer power. A common form is K(x, z) = (1 + x dot z)^d.
Polynomial matrix spectral factorization - Polynomial matrix spectral factorization expresses a suitable matrix-valued spectrum as a polynomial matrix multiplied by its conjugate-transposed...
Polynomial regression - Polynomial regression models an outcome using powers of one or more numerical predictors.
Polynomial trend coefficient - A polynomial trend coefficient is a numerical weight used with other such weights to test a specified trend across ordered group means.
Polysemanticity - Polysemanticity is the property of a neural-network component, often an individual neuron, responding to several distinct features or concepts rather than...
POMCP - POMCP stands for Partially Observable Monte Carlo Planning.
Pooled variance - Pooled variance is an estimate formed by combining variance information from several samples.
Pooling - Pooling is an operation that summarizes a group of values into a smaller set of representative values.
Pooling Layer - A pooling layer summarizes groups of values in a neural-network representation.
Popularity Bias - Popularity bias is a tendency for already popular items to receive disproportionate exposure while less popular items receive little representation.
Population - In statistics, a population is the complete set of units, events or measurements about which a question is asked.
Population Bias - Population bias is a systematic mismatch between the people represented in a dataset and the population about which conclusions or predictions are intended.
Population protocol - A population protocol is a model of distributed computation in which many simple agents interact in pairs and update their states according to shared rules.
Population variance - Population variance measures the average squared distance of values from their population mean.
Pose - Pose is an object's position and orientation relative to a specified reference frame. It describes both where the object is and how it is turned.
Positional embeddings - Positional embeddings are numerical representations that supply information about where tokens or other elements occur in a sequence.
Positional encoding - Positional encoding is the information added to a transformer's inputs so it can tell where each token sits in the sequence, because the architecture itself...
Positional index - A positional index is an inverted search index that records where each term occurs within each document, usually as token positions.
Positive class - The positive class is the category a classification task designates as the condition or event being detected.
Positive definite (pd) - A real symmetric matrix is positive definite, abbreviated pd, when its quadratic form is strictly positive for every nonzero vector.
Positive semi-definite (psd) - A real symmetric matrix is positive semi-definite, abbreviated psd, when its quadratic form is nonnegative for every vector.
Positively skewed - A positively skewed distribution is asymmetric with a tail extending toward larger values, conventionally shown to the right on a horizontal axis.
Possessive pronoun - A possessive pronoun is a pronominal expression that indicates a possession-like relationship without repeating a full possessor expression.
Possibility theory - Possibility theory is a mathematical framework for representing uncertainty through degrees of possibility and necessity.
Possible world - A possible world is a specified way a situation could be, used to evaluate whether statements are true or false.
Post-editing - Post-editing is the revision of machine-generated output after it has been produced.
Post hoc analysis - Post hoc analysis is analysis undertaken after results have been examined, often to explore an unexpected pattern or investigate a broader finding in more...
Post-hoc explanation - A post-hoc explanation is an explanation produced by analyzing an already developed AI model or its behavior.
Post-processing - Post-processing is the adjustment or transformation of a model's output after inference.
Post-processing algorithm - A post-processing algorithm is a defined procedure that transforms a model's outputs after prediction.
Post-sample evaluation - Post-sample evaluation assesses a forecasting model using observations from after the period on which it was estimated.
Post-trained model - A post-trained model is a model that has undergone additional adaptation or refinement after an initial pre-training stage.
Post-training - Post-training is additional training applied to an already pretrained model to shape its behavior or capabilities.
Posterior - A posterior is a probability distribution updated using observed evidence within a specified model.
Posterior (prediction) - In the prediction sense used here, a posterior is the conditional probability distribution of an unknown label or target given observed features.
Posterior predictive check - A posterior predictive check compares observed data with replicated data generated from a fitted Bayesian model.
Posterior predictive distribution - A posterior predictive distribution describes possible future or replicated observations after accounting for observed data through a Bayesian model.