AI Tasks

Representing the key concepts and relations between the decision variables in some formal manner, typically within a framework suggested by an expert systems shell, i.e. encoding of knowledge in a form that can be used for computer-based problem solving.  Knowledge: Awareness or familiarity gained by experiences of facts, data, and situations.  The kind of knowledge…

Constraint satisfaction problems are closely related to optimization problems. The space of possible solutions to a constraint satisfaction problem is typically defined by a set of decision variables (e.g. the squares to be filled in Sudoku). The task is to find some particular assignment to these variables that will meet a predefined set of constraints.…

Optimization is one of the key topics in the area of artificial intelligence and machine learning. Especially in the sense that the entire field of machine learning is based on optimization, but also in the sense that some artificial intelligence and machine learning methods can do optimization. Loosely speaking, optimization is the process of finding…

Planning can be defined as “the reasoning side of acting” [auto_planning]. Solving a planning problem involves thinking about the available actions, considering their expected effects and identifying a sequence of them that will achieve the desired goal. Planning problems can be specified in multiple ways. Often there is the concept of a goal state that…

The task of cluster analysis involves identifying clusters (groups) of points in a dataset. The principal characteristic of a cluster is that its points are much closer to each other (in some sense: usually based on a well-defined distance measure) than they are to points outside the cluster. Cluster analysis: an illustration. Cluster analysis (or…

A specific class of problems within the domain of data analysis and often associated with supervised supervised learning tasks such as classification and regression is collectively known as time series analysis and its distinguishing feature is that it is concerned with temporal data. By time series we mean a series of data, where each item…

Classification is a task of supervised learning: a classifier is fitted to map its inputs (which can be discrete or continuous) to classes. The classifier observes an instance and decides into which of a predefined number of classes it belongs. Problems of this kind may include problems with well-structured data such as predicting whether a…

Regression is a task of supervised learning: a regression model is fitted to map its inputs (which can be discrete or continuous) to some previously known outputs. The outputs of a regression model are continuous. Practical problems of this kind may include tasks such as predicting real-estate prices, predicting demand for a service or a…

Visual semantic segmentation is a task related to, but distinct from image recognition and visual object detection. Like visual object detection and unlike image recognition, a segmentation method provides information about where an object occurs in the image. However, in comparison to visual object detection, in segmentation, annotations are much more fine-grained: they come at…

Identification – especially when based on unstructured data such as audio or vision – is a problem that often requires sophisticated machine learning methods. The problem of identification is distinct from both image recognition and visual object detection. Where image recognition and visual object detection work on classes of images, identification deals with particular instances.…

The problem of visual object detection, as generally understood, is distinct from image recognition. In image recognition, an image as a whole is classified into one (or several) of a number of classes. In visual detection, on the other hand, the detector picks up the one or several instances of an object in the image…

Modeling is a core topic across many disciplines. Generally speaking, a model is a simplified representation of some system, which only captures the aspects and characteristics of the original system that are important for a particular purpose. The relative importance of the various aspects depends, of course, on the intended use of the model. Artificial…

The area of decision making and control involves selecting among several different courses of action. The available actions can be: Discrete: finite number; infinite number; Continuous: infinite number. The decisions themselves can be: One-time: a single decision with different possible outcomes; Sequential: a sequence of decisions affecting the state of the same environment; a.k.a. discrete…