arXiv subjects

Imre Kiss

Publications and source records attributed to Imre Kiss.

5 recordsLinked to original sources

arXiv2020

WISE: A Computer System Performance Index Scoring Framework

The performance levels of a computing machine running a given workload configuration are crucial for both users and providers of computing resources. Knowing how well a computing machine is running with a given workload configuration is critical to making proper computing resource allocation decisions. In this paper, we introduce a novel framework for deriving computing machine and computing resource performance indicators for a given workload configuration. We propose a workload/machine index score (WISE) framework for computing a fitness score for a workload/machine combination. The WISE score indicates how well a computing machine is running with a specific workload configuration by addressing the issue of whether resources are being stressed or sitting idle wasting precious resources. In addition to encompassing any number of computing resources, the WISE score is determined by considering how far from target levels the machine resources are operating at without maxing out. Experimental results demonstrate the efficacy of the proposed WISE framework on two distinct workload configurations.

cs.DC↗

arXiv2020

Self-Attention Gazetteer Embeddings for Named-Entity Recognition

Recent attempts to ingest external knowledge into neural models for named-entity recognition (NER) have exhibited mixed results. In this work, we present GazSelfAttn, a novel gazetteer embedding approach that uses self-attention and match span encoding to build enhanced gazetteer embeddings. In addition, we demonstrate how to build gazetteer resources from the open source Wikidata knowledge base. Evaluations on CoNLL-03 and Ontonotes 5 datasets, show F1 improvements over baseline model from 92.34 to 92.86 and 89.11 to 89.32 respectively, achieving performance comparable to large state-of-the-art models.

cs.CL↗

arXiv2019

Active Learning for New Domains in Natural Language Understanding

We explore active learning (AL) for improving the accuracy of new domains in a natural language understanding (NLU) system. We propose an algorithm called Majority-CRF that uses an ensemble of classification models to guide the selection of relevant utterances, as well as a sequence labeling model to help prioritize informative examples. Experiments with three domains show that Majority-CRF achieves 6.6%-9% relative error rate reduction compared to random sampling with the same annotation budget, and statistically significant improvements compared to other AL approaches. Additionally, case studies with human-in-the-loop AL on six new domains show 4.6%-9% improvement on an existing NLU system.

cs.CL↗

arXiv2019

F10-SGD: Fast Training of Elastic-net Linear Models for Text Classification and Named-entity Recognition

Voice-assistants text classification and named-entity recognition (NER) models are trained on millions of example utterances. Because of the large datasets, long training time is one of the bottlenecks for releasing improved models. In this work, we develop F10-SGD, a fast optimizer for text classification and NER elastic-net linear models. On internal datasets, F10-SGD provides 4x reduction in training time compared to the OWL-QN optimizer without loss of accuracy or increase in model size. Furthermore, we incorporate biased sampling that prioritizes harder examples towards the end of the training. As a result, in addition to faster training, we were able to obtain statistically significant accuracy improvements for NER. On public datasets, F10-SGD obtains 22% faster training time compared to FastText for text classification. And, 4x reduction in training time compared to CRFSuite OWL-QN for NER.

cs.CL↗

arXiv2010

Steepest Ascent Hill Climbing For A Mathematical Problem

The paper proposes artificial intelligence technique called hill climbing to find numerical solutions of Diophantine Equations. Such equations are important as they have many applications in fields like public key cryptography, integer factorization, algebraic curves, projective curves and data dependency in super computers. Importantly, it has been proved that there is no general method to find solutions of such equations. This paper is an attempt to find numerical solutions of Diophantine equations using steepest ascent version of Hill Climbing. The method, which uses tree representation to depict possible solutions of Diophantine equations, adopts a novel methodology to generate successors. The heuristic function used help to make the process of finding solution as a minimization process. The work illustrates the effectiveness of the proposed methodology using a class of Diophantine equations given by a1. x1 p1 + a2. x2 p2 + ...... + an . xn pn = N where ai and N are integers. The experimental results validate that the procedure proposed is successful in finding solutions of Diophantine Equations with sufficiently large powers and large number of variables.

cs.AI↗