Editing Tabula rasa (section)

===Computer science===
In [[artificial intelligence]], ''tabula rasa'' refers to the development of autonomous agents with a mechanism to reason and plan toward their goal, but no "built-in" knowledge-base of their environment. Thus, they truly are a blank slate.{{cn|date=August 2024}}

In reality, autonomous agents possess an initial [[Data set|data-set]] or knowledge-base, but this cannot be immutable or it would hamper autonomy and heuristic ability. Even if the data-set is empty, it usually may be argued that there is a built-in [[bias]] in the reasoning and planning mechanisms. Either intentionally or unintentionally placed there by the human designer, it thus negates the true spirit of ''tabula rasa''.<ref>[http://www.catb.org/jargon/html/koans.html The Jargon Files: "Sussman attains enlightenment"], also see the article section [[Hacker koan#Uncarved block|Hacker koan: Uncarved block]]</ref>

A synthetic (programming) language [[Parsing|parser]] ([[LR(1)]], [[LALR(1)]] or [[SLR(1)]], for example) could be considered a special case of a ''tabula rasa'', as it is designed to accept ''any'' of a possibly infinite set of source language programs, within a ''single'' programming language, and to output either a good parse of the program, or a good machine language translation of the program, either of which represents a ''success'', or, alternately, a ''failure'', and nothing else. The "initial data-set" is a set of tables which are generally produced mechanically by a parser table generator, usually from a [[Backus–Naur form|BNF]] representation of the source language, and represents a "table representation" of that ''single'' programming language.

[[AlphaZero]] achieved superhuman performance in [[chess]] and [[shogi]] using [[Self-play (reinforcement learning technique)|self-play]] and ''tabula rasa'' [[reinforcement learning]], meaning it had no access to human games or [[Hard coding|hard-coded]] human knowledge about either board game, only the rules of the games.<ref name="preprint">Silver, David, Thomas Hubert, Julian Schrittwieser, Ioannis Antonoglou, et al. 2017. "[https://arxiv.org/pdf/1712.01815.pdf Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm]." {{Arxiv|1712.01815}} [https://arxiv.org/archive/cs.AI cs.AI].</ref>