Types of monotonic language learning and their characterization

Lange, Steffen; Zeugmann, Thomas

doi:10.1145/130385.130427

Cited by 69 publications

(31 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This contrasts with the situation for learning from texts, where every indexed family which can be strongly monotonically identified from texts can be strongly monotonically identified by an iterative learner from texts (using some appropriate hypothesis space) [LZ92].…”

Section: Resultsmentioning

confidence: 93%

“…We show that there exists an indexed family which can be strongly monotonically identified using a class preserving hypothesis space but cannot be strongly monotonically identified by an iterative learner, even if the iterative learner is free in the choice of the hypothesis space used. This contrasts with the situation for learning from texts, where every indexed family which can be strongly monotonically identified can also be strongly monotonically identified by an iterative learner using some appropriate hypothesis space [LZ92].…”

Section: Introductionmentioning

confidence: 88%

“…We refer the reader to [LZ96] for a recent survey on learning indexed families. Lange and Zeugmann [LZ92] asked whether there exists an indexed family which can be strongly monotonically identified by a learner from informant, but which cannot be strongly monotonically identified by an iterative learner from informant. We show that there exists an indexed family which can be strongly monotonically identified using a class preserving hypothesis space but cannot be strongly monotonically identified by an iterative learner, even if the iterative learner is free in the choice of the hypothesis space used.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

On some open problems in monotonic and conservative learning

Jain¹

2009

Information Processing Letters

View full text Add to dashboard Cite

show abstract

Section: Resultsmentioning

confidence: 93%

Section: Introductionmentioning

confidence: 88%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

On some open problems in monotonic and conservative learning

Jain¹

2009

Information Processing Letters

View full text Add to dashboard Cite

show abstract

“…(This latter subset principle, for preventing overgeneralization, is further discussed, for example, in [8,12,13,24,34,3,35,23]. Mukouchi [27] and Lange and Zeugmann [25] present a subset principle for one-shot learning. ) Even at the TxtBc levels, noise is problematic.…”

Section: Disadvantages Of Having Noise In the Inputmentioning

confidence: 99%

Vacillatory and BC learning on noisy data

Case¹,

Jain²,

Stephan³

1996

Lecture Notes in Computer Science

View full text Add to dashboard Cite

The present work employs a model of noise introduced earlier by the third author. In this model noisy data nonetheless uniquely determines the true data: correct information occurs infinitely often while incorrect information occurs only finitely often. The present paper considers the effects of this form of noise on vacillatory and behaviorally correct learning of grammars -both from positive data alone and from informant (positive and negative data). For learning from informant, the noise, in effect, destroys negative data. Various noisy-data hierarchies are exhibited, which, in some cases, are known to collapse when there is no noise. Noisy behaviorally correct learning is shown to obey a very strong "subset principle". It is shown, in many cases, how much power is needed to overcome the effects of noise. For example, the best we can do to simulate, in the presence of noise, the noise-free, no mind change cases takes infinitely many mind changes. One technical result is proved by a priority argument.

show abstract

“…In Wiehagen (1976) CON S ⊂ IT ⊂ LIM has been proved. Recent papers give new evidence for the power of iterative learning (cf., e.g., Porat and Feldman (1988), Lange and Wiehagen (1991), Lange and Zeugmann (1992)). …”

Section: The General Inconsistency Phenomenonmentioning

confidence: 99%

Learning and consistency

Wiehagen

Zeugmann

1995

Algorithmic Learning for Knowledge-Based Systems

Self Cite

View full text Add to dashboard Cite

In designing learning algorithms it seems quite reasonable to construct them in such a way that all data the algorithm already has obtained are correctly and completely reflected in the hypothesis the algorithm outputs on these data. However, this approach may totally fail. It may lead to the unsolvability of the learning problem, or it may exclude any efficient solution of it. Therefore we study several types of consistent learning in recursion-theoretic inductive inference. We show that these types are not of universal power. We give "lower bounds" on this power. We characterize these types by some versions of decidability of consistency with respect to suitable "non-standard" spaces of hypotheses. Then we investigate the problem of learning consistently in polynomial time. In particular, we present a natural learning problem and prove that it can be solved in polynomial time if and only if the algorithm is allowed to work inconsistently.

show abstract

Types of monotonic language learning and their characterization

Abstract: The present paper deals with strong-monoto-

Cited by 69 publications

References 13 publications

On some open problems in monotonic and conservative learning

On some open problems in monotonic and conservative learning

Vacillatory and BC learning on noisy data

Learning and consistency

Contact Info

Product

Resources

About