Ticker: A system for incremental ASP-based stream reasoning

Abstract

In complex reasoning tasks, as expressible by Answer Set Programming (ASP), problems often permit for multiple solutions. In dynamic environments, where knowledge is continuously changing, the question arises how a given model can be incrementally adjusted relative to new and outdated information. This paper introduces Ticker, a prototypical engine for well-defined logical reasoning over streaming data. Ticker builds on a practical fragment of the recent rule-based language LARS, which extends ASP for streams by providing flexible expiration control and temporal modalities. We discuss Ticker's reasoning strategies: first, the repeated one-shot solving mode calls Clingo on an ASP encoding. We show how this translation can be incrementally updated when new data is streaming in or time passes by. Based on this, we build on Doyle's classic justification-based truth-maintenance system to update models of non-stratified programs. Finally, we empirically compare the obtained evaluation mechanisms.

Reasoning in Highly Reactive Environments

“…In the latest years, due to the increasing volume of data streams and the crucial time requirements of many applications, different real-time approaches have been studied in order to ensure that a query answer is pushed as soon as it becomes available to the consumers. Notably, in the field of stream reasoning, some work aim at introducing advanced reasoning capabilities under a well understood and formalized framework [3,4]. By the way, Stream Reasoning is a quite young and unexplored area, and therefore many different issues need to be solved.…”

“…In the context of stream reasoning and multi-shot reasoning engines [4,6,16] based on answer set programming, a quite typical setting is when the grounding step is repeatedly executed on slightly different input data, while a short computation time window is allowed. In this respect, we decided to focus our attention on the following open issues:…”

“…In order to achieve our goal, we started to investigate the literature [1,3,4,7,14,15,17,19,21,22]. Then, we focused our research on artificial agents for video-games (also called "bots").…”

“…As a side research topic we investigated to what extent the ASP-based approach is scalable enough for industrial contexts in the field of video games, by proposing a Unity extension capable to automatically generate dungeons maps 4 . In this context we first investigated over the usage of a partition-based generation technique [25], then we proposed a multiple step-generation approach, set in the context of the 2-D caves generation domain, where each step is declaratively controlled by an ASP specification.…”

Expressive Stream Reasoning with Laser

“…Then, we can add to Alg. 1 an additional for-loop around lines [6][7][8][9][10][11][12][13] to compute the answer stream stratum by stratum bottom up as usual. In fact, our implementation makes use of this extension.…”