Inferring Grammars for Mildly Context Sensitive Languages in Polynomial-Time

Bio-inspired Grammatical Inference

“…We studied a mechanism that have such interesting properties, called Simple External Contextual. Our main results can be found in [9,25,6]. …”

Efficient learning of multiple context-free languages with multidimensional substitutability from positive data

“…For instance, ⟨a, b c d, e ⟩ ⊙ ⟨y 1 , y 2 , y 3 , y 4 ⟩ = ⟨a, by 1 cy 2 d, y 3 ey 4 ⟩. Simple external contextual (sec) languages are important mildly context-sensitive languages in the context of grammatical inference [4,19,2]. For p ≥ 1 and q ≥ 0, a p, q-sec grammar G over Σ is a pair ⟨B, C ⟩ where B ∈ (Σ * ) ⟨p⟩ and C ⊆ (Σ * Σ * ) ⟨p⟩ with |C| ≤ q.…”

“…Their work has led to several fruitful results in grammatical inference [7,11,25], which target even larger classes of context-free languages with some special properties related to substitutability. In addition, now, mildly context-sensitive languages have arisen as a topical target of grammatical inference [4,19,3,16]. Since some non-context-free phenomena, like cross-serial dependencies in Dutch and Swiss-German [5,23], had been found in natural languages, the notion of mildly context-sensitive languages was proposed for better describing natural languages while keeping tractability [14].…”

Children as Models for Computers: Natural Language Acquisition for Machine Learning

“…The learning algorithm derived from their main result was not time efficient. In [13], it was presented a polynomial-time algorithm for inferring SEC from positive data (small values of p and q were considered). Later, in [14], it was investigated for which choice of the parameter q (denoting the number of contexts) the class of SEC is iteratively learnable.…”