Untitled

Vilalta, Ricardo; Drissi, Youssef

doi:10.1023/a:1019956318069

Cited by 808 publications

(126 citation statements)

References 45 publications

Supporting

Mentioning

103

Contrasting

Order By: Relevance

“…This last result may reveal very interesting, as it can suggest that a second layer of (meta) algorithm selection could maximize the performance of the first. Such result meets the insights of [26] in the suggestion of a recursion of adaptive learners as a possible new paradigm of meta-learning.…”

Section: Resultssupporting

confidence: 75%

See 1 more Smart Citation

Meta-mining Evaluation Framework: A Large Scale Proof of Concept on Meta-learning

Raynaut

Soulé-Dupuy

Vallès-Parlangeau

2016

AI 2016: Advances in Artificial Intelligence

View full text Add to dashboard Cite

OATAO is an open access repository that collects the work of Toulouse researchers and makes it freely available over the web where possible. This is an author-deposited version published in : http://oatao.univ-toulouse.fr/ Eprints ID : 18799The contribution was presented at AI 2016 :https://ai2016.net/ Abstract. This paper aims to provide a unified framework for the evaluation and comparison of the many emergent meta-mining techniques. This framework is illustrated on the case study of the meta-learning problem in a large scale experiment. The results of this experiment are then explored through hypothesis testing in order to provide insight regarding the performance of the different meta-learning schemes, advertising the potential of our approach regarding meta-level knowledge discovery.

show abstract

Section: Resultssupporting

confidence: 75%

“…This approach, notably appearing in the STATLOG project [9,13], and in most studies afterwards [7,26], allows the use of many expressive measures. But its performance depends heavily on the adequateness of bias between the meta-level learner and the chosen measures.…”

Section: Dimensions Of Studymentioning

confidence: 99%

Meta-mining Evaluation Framework: A Large Scale Proof of Concept on Meta-learning

Raynaut

Soulé-Dupuy

Vallès-Parlangeau

2016

AI 2016: Advances in Artificial Intelligence

View full text Add to dashboard Cite

show abstract

“…A number of definitions of metalearning have been given, the following list cites the main review papers and books from the last decade:Metalearning studies how learning systems can increase in efficiency through experience; the goal is to understand how learning itself can become flexible according to the domain or task under study (Vilalta and Drissi 2002a). …”

Section: Definitionmentioning

confidence: 99%

“…Metalearning studies how learning systems can increase in efficiency through experience; the goal is to understand how learning itself can become flexible according to the domain or task under study (Vilalta and Drissi 2002a). …”

Section: Definitionmentioning

confidence: 99%

“…Dynamic bias selection is often mentioned in relation to a continuous flow of training examples (data streams), where metalearning algorithms can be used for bias management and detecting concept drifts, as argued in Brazdil et al (2009). In their metalearning review, Vilalta and Drissi (2002a) cite Bruha and Famili (2000) as the only example of dynamic bias selection apart from a few projects before 1992. In this contribution, a rule-based system which includes quality information about each rule, influencing the prediction process is used.…”

Section: Notions Of Metalearningmentioning

confidence: 99%

See 1 more Smart Citation

Metalearning: a survey of trends and technologies

Lemke¹,

2013

View full text Add to dashboard Cite

Metalearning attracted considerable interest in the machine learning community in the last years. Yet, some disagreement remains on what does or what does not constitute a metalearning problem and in which contexts the term is used in. This survey aims at giving an all-encompassing overview of the research directions pursued under the umbrella of metalearning, reconciling different definitions given in scientific literature, listing the choices involved when designing a metalearning system and identifying some of the future research challenges in this domain.

show abstract

Research Activities on Magnesium Alloys at Seoul National University, Korea

Shin

2003

Magnesium

View full text Add to dashboard Cite

We propose a method of quantum machine learning called quantum mutual information neural estimation (QMINE) for estimating von Neumann entropy and quantum mutual information, which are fundamental properties in quantum information theory. The QMINE proposed here basically utilizes a technique of quantum neural networks (QNNs), to minimize a loss function that determines the von Neumann entropy, and thus quantum mutual information, which is believed more powerful to process quantum datasets than conventional neural networks due to quantum superposition and entanglement. To create a precise loss function, we propose a quantum Donsker-Varadhan representation (QDVR), which is a quantum analog of the classical Donsker-Varadhan representation. By exploiting a parameter shift rule on parameterized quantum circuits, we can efficiently implement and optimize the QNN and estimate the quantum entropies using the QMINE technique. Furthermore, numerical observations support our predictions of QDVR and demonstrate the good performance of QMINE.

show abstract

Untitled

Cited by 808 publications

References 45 publications

Meta-mining Evaluation Framework: A Large Scale Proof of Concept on Meta-learning

Meta-mining Evaluation Framework: A Large Scale Proof of Concept on Meta-learning

Metalearning: a survey of trends and technologies

Research Activities on Magnesium Alloys at Seoul National University, Korea

Contact Info

Product

Resources

About