Cache-based Dynamic PCFG Adaptation using MAP Estimation
ثبت نشده
چکیده
This paper presents a cache-based dynamic adaptation technique for lexicalized probabilistic context-free-grammar (LPCFG). Expected counts from machine-parsed sentences of in-domain data are stored in a cache, which are combined with prior counts from hand-annotated parses of outof-domain data using maximum a posteriori (MAP) estimation. This adaptation is unsupervised, and dynamic with an adaptation model created at runtime. Experiments on the Tsinghua Chinese Treebank (a balanced treebank with different domains) show that our adaptation techniques provide a significant improvement as much as .6% average F1 measure.
منابع مشابه
Speaker Adaptation in Continuous Speech Recognition Using MLLR-Based MAP Estimation
A variety of methods are used for speaker adaptation in speech recognition. In some techniques, such as MAP estimation, only the models with available training data are updated. Hence, large amounts of training data are required in order to have significant recognition improvements. In some others, such as MLLR, where several general transformations are applied to model clusters, the results ar...
متن کاملSpeaker Adaptation in Continuous Speech Recognition Using MLLR-Based MAP Estimation
A variety of methods are used for speaker adaptation in speech recognition. In some techniques, such as MAP estimation, only the models with available training data are updated. Hence, large amounts of training data are required in order to have significant recognition improvements. In some others, such as MLLR, where several general transformations are applied to model clusters, the results ar...
متن کاملSupervised and unsupervised PCFG adaptation to novel domains
This paper investigates adapting a lexicalized probabilistic context-free grammar (PCFG) to a novel domain, using maximum a posteriori (MAP) estimation. The MAP framework is general enough to include some previous model adaptation approaches, such as corpus mixing in Gildea (2001), for example. Other approaches falling within this framework are more effective. In contrast to the results in Gild...
متن کاملCache Replacement Policy Using Map-based Adaptive Insertion
In this paper, we propose a map-based adaptive insertion policy (MAIP) for a novel cache replacement. The MAIP estimates the data reuse possibility on the basis of data reuse history. To track data reuse history, the MAIP employs a bitmap data structure, which we call memory access map. The memory access map holds all memory accessed locations in a fixed sized memory area to detect the data reu...
متن کاملContext Adaptation in Statistical Machine Translation Using Models with Exponentially Decaying Cache
We report results from a domain adaptation task for statistical machine translation (SMT) using cache-based adaptive language and translation models. We apply an exponential decay factor and integrate the cache models in a standard phrasebased SMT decoder. Without the need for any domain-specific resources we obtain a 2.6% relative improvement on average in BLEU scores using our dynamic adaptat...
متن کامل