2014 AnIntroductiontoConstraintBased
- (Barták et al., 2014) ⇒ Roman Barták, Robert A Morris, and K Brent Venable. (2014). “An Introduction to Constraint-Based Temporal Reasoning.” In: Synthesis Lectures on Artificial Intelligence and Machine Learning Journal, 8(1). doi:10.2200/S00557ED1V01Y201312AIM026
Subject Headings: Constraint-based Temporal Reasoning, Temporal Reasoning, Constraint-based Reasoning.
Notes
Cited By
Quotes
Abstract
Solving challenging computational problems involving time has been a critical component in the development of artificial intelligence systems almost since the inception of the field. This book provides a concise introduction to the core computational elements of temporal reasoning for use in AI systems for planning and scheduling, as well as systems that extract temporal information from data. It presents a survey of temporal frameworks based on constraints, both qualitative and quantitative, as well as of major temporal consistency techniques. The book also introduces the reader to more recent extensions to the core model that allow AI systems to explicitly represent temporal preferences and temporal uncertainty. This book is intended for students and researchers interested in constraint-based temporal reasoning. It provides a self-contained guide to the different representations of time, as well as examples of recent applications of time in AI systems.
1. Introduction to Time in AI Systems
Physics teaches us that time is fundamental to the understanding of the material world. Continuous physical processes like waves, flows, and oscillations are characterized in terms of changes, and rates of change, of the world over time.
Time provides a mental substrate for the human management of perception and action. In particular, time provides a necessary cognitive and linguistic component for describing change. Change happens through the occurrence of events, processes, and actions, and time provides a way to record, order, and measure the duration of these occurrences. Indeed, periodic events and motion have long served as standards for units of time. Currently, the international unit of time, the second, is defined in terms of radiation emitted by caesium atoms. [118]
Time management is a fundamental aspect of intelligent behavior. Time management consists of describing, predicting, and planning actions or events. Artifacts for time management such as sundials, mechanical clocks, and calendars, arose from the need to record and predict when events or natural processes recur. The human organization and visualization of time in linear form, as points on a line, has ancient origins [96]. But it was primarily the modern industrial age that gave rise to the human awareness of what makes time management mentally complex. Computing machines arose as a natural technology to apply to solve problems in time management. In this chapter, we briefly trace the development of time management problems, their formulation in mathematics and logic, and the rise of algorithmic solutions, including those based in AI. We conclude this chapter with an overview of the goals and scope of this book.
…
2. Temporal Frameworks Based on Constraints
The representation of temporal information, and reasoning about time, are important in artificial intelligence. Reasoning about time plays an important role in building automated planning and scheduling systems, where causal and temporal relations are the most critical concepts. Time also enters into areas such as natural language processing, for example to represent stories. In these areas we need to express information about events and processes such as “reading newspapers” and “having a breakfast” that span over an interval of time and that are related to each other, as expressed in sentences such as “I read the newspaper during breakfast.”
A time-aware rational agent abstracts information about past events and observed processes from the sensors. These temporal references (time points and intervals) and temporal relations between them (such as “some event happened before another event” and “a given process had a specific duration”) are stored in a temporal knowledge base. To do reasoning such as activity planning, the agent adds temporal references about future events (goals, etc.) and futures processes (planned activities) and asks if all these temporal references and relations are consistent. This way the agent infers what may happen in future, what is inevitable, and what is impossible. More formally, any system that contains an explicit representation of time contains temporal references (time points and intervals) and temporal propositions describing the temporal relations between the temporal references. These components combine into a temporal reasoning system consisting of
- a temporal knowledge base containing temporal propositions,
- a procedure for checking consistency of the temporal propositions, and
- an inference mechanism that is used to deduce new information and answer queries about the temporal references.
In this chapter we will describe the core frameworks for representing temporal references and for expressing temporal propositions. We will also introduce the basic algorithms for consistency checking and we will survey the complexity results. We will abstract from the mechanisms to obtain temporal information (from sensors, etc.) and to use it in further reasoning (such as activity planning). The focus will be on formal models of time and on temporal reasoning in these models.
…
3. Extensions: Preferences and Uncertainty
4. Applications of Temporal Reasoning
References
;
| Author | volume | Date Value | title | type | journal | titleUrl | doi | note | year | |
|---|---|---|---|---|---|---|---|---|---|---|
| 2014 AnIntroductiontoConstraintBased | Robert A Morris K Brent Venable Roman Barták | An Introduction to Constraint-Based Temporal Reasoning | 10.2200/S00557ED1V01Y201312AIM026 |