Kashif Munir - Academia.edu (original) (raw)
Papers by Kashif Munir
ACM Transactions on Asian and Low-Resource Language Information Processing
The recent success of neural networks in NLP applications has provided a strong impetus to develo... more The recent success of neural networks in NLP applications has provided a strong impetus to develop supervised models for semantic role labeling (SRL) that forego the requirement for extensive feature engineering. Recent state-of-the-art approaches require high-quality annotated datasets that are costly to obtain and almost unavailable for low-resource languages. We present a semi-supervised approach that utilizes both labeled and unlabeled data to provide performance improvement over a mere supervised SRL model. We show that our proposed semi-supervised SRL model provides larger improvement over a supervised model in the scenario where labeled training data size is small. Our SRL system leverages unlabeled data under the language modeling paradigm. We demonstrate that the incorporation of a self pre-trained bidirectional language model (S-PrLM) into a SRL system can help in SRL performance improvement by learning composition functions from the unlabeled data. Previous researches hav...
ACM Transactions on Asian and Low-Resource Language Information Processing
The task of semantic role labeling ( SRL ) is dedicated to finding the predicate-argument structu... more The task of semantic role labeling ( SRL ) is dedicated to finding the predicate-argument structure. Previous works on SRL are mostly supervised and do not consider the difficulty in labeling each example which can be very expensive and time-consuming. In this article, we present the first neural unsupervised model for SRL. To decompose the task as two argument related subtasks, identification and clustering, we propose a pipeline that correspondingly consists of two neural modules. First, we train a neural model on two syntax-aware statistically developed rules. The neural model gets the relevance signal for each token in a sentence, to feed into a BiLSTM, and then an adversarial layer for noise-adding and classifying simultaneously, thus enabling the model to learn the semantic structure of a sentence. Then we propose another neural model for argument role clustering, which is done through clustering the learned argument embeddings biased toward their dependency relations. Experim...
IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2021
Semantic role labeling (SRL) aims at elaborating the meaning of a sentence by forming a predicate... more Semantic role labeling (SRL) aims at elaborating the meaning of a sentence by forming a predicate-argument structure. Recent researches depicted that the effective use of syntax can improve SRL performance. However, syntax is a complicated linguistic clue and is hard to be effectively applied in a downstream task like SRL. This work effectively encodes syntax using adaptive convolution which endows strong flexibility to existing convolutional networks. The existing CNNs may help in encoding a complicated structure like syntax for SRL, but it still has shortcomings. Contrary to traditional convolutional networks that use same filters for different inputs, adaptive convolution uses adaptively generated filters conditioned on syntacticallyinformed inputs. We achieve this with the integration of a filter generation network which generates the input specific filters. This helps the model to focus on important syntactic features present inside the input, thus enlarging the gap between syntax-aware and syntax-agnostic SRL systems. We further study a hashing technique to compress the size of the filter generation network for SRL in terms of trainable parameters. Experiments on CoNLL-2009 dataset confirm that the proposed model substantially outperforms most previous SRL systems for both English and Chinese languages.
2015 IEEE 15th International Conference on Environment and Electrical Engineering (EEEIC), 2015
ABSTRACT Demand of power is increasing worldwide and this increased demand has motivated the engi... more ABSTRACT Demand of power is increasing worldwide and this increased demand has motivated the engineers to look for alternative or renewable energy systems. Meanwhile, the data centers are also employing the solar energy systems to avoid the AC/DC power conversion losses due to the fact that an efficient solar system with DC/DC converter will help them to save up to 15% of the power in comparison to the conventional power supply system using non-renewable energy sources. The efficiency of the solar energy systems is greatly dependent on the DC/DC converters and hence, there is a need to study the comparison of efficiency of such converters which can be used in Switch Model Power Supplies (SMPS). This paper provides a comparative analysis between some of the DC/DC converters i.e. synchronous buck converter, isolated buck converter and non-isolated buck converter. Moreover, the important factors such as input voltage variation and the effects of open and closed loop systems of mentioned DC/DC converters are also explained.
ACM Transactions on Asian and Low-Resource Language Information Processing
The recent success of neural networks in NLP applications has provided a strong impetus to develo... more The recent success of neural networks in NLP applications has provided a strong impetus to develop supervised models for semantic role labeling (SRL) that forego the requirement for extensive feature engineering. Recent state-of-the-art approaches require high-quality annotated datasets that are costly to obtain and almost unavailable for low-resource languages. We present a semi-supervised approach that utilizes both labeled and unlabeled data to provide performance improvement over a mere supervised SRL model. We show that our proposed semi-supervised SRL model provides larger improvement over a supervised model in the scenario where labeled training data size is small. Our SRL system leverages unlabeled data under the language modeling paradigm. We demonstrate that the incorporation of a self pre-trained bidirectional language model (S-PrLM) into a SRL system can help in SRL performance improvement by learning composition functions from the unlabeled data. Previous researches hav...
ACM Transactions on Asian and Low-Resource Language Information Processing
The task of semantic role labeling ( SRL ) is dedicated to finding the predicate-argument structu... more The task of semantic role labeling ( SRL ) is dedicated to finding the predicate-argument structure. Previous works on SRL are mostly supervised and do not consider the difficulty in labeling each example which can be very expensive and time-consuming. In this article, we present the first neural unsupervised model for SRL. To decompose the task as two argument related subtasks, identification and clustering, we propose a pipeline that correspondingly consists of two neural modules. First, we train a neural model on two syntax-aware statistically developed rules. The neural model gets the relevance signal for each token in a sentence, to feed into a BiLSTM, and then an adversarial layer for noise-adding and classifying simultaneously, thus enabling the model to learn the semantic structure of a sentence. Then we propose another neural model for argument role clustering, which is done through clustering the learned argument embeddings biased toward their dependency relations. Experim...
IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2021
Semantic role labeling (SRL) aims at elaborating the meaning of a sentence by forming a predicate... more Semantic role labeling (SRL) aims at elaborating the meaning of a sentence by forming a predicate-argument structure. Recent researches depicted that the effective use of syntax can improve SRL performance. However, syntax is a complicated linguistic clue and is hard to be effectively applied in a downstream task like SRL. This work effectively encodes syntax using adaptive convolution which endows strong flexibility to existing convolutional networks. The existing CNNs may help in encoding a complicated structure like syntax for SRL, but it still has shortcomings. Contrary to traditional convolutional networks that use same filters for different inputs, adaptive convolution uses adaptively generated filters conditioned on syntacticallyinformed inputs. We achieve this with the integration of a filter generation network which generates the input specific filters. This helps the model to focus on important syntactic features present inside the input, thus enlarging the gap between syntax-aware and syntax-agnostic SRL systems. We further study a hashing technique to compress the size of the filter generation network for SRL in terms of trainable parameters. Experiments on CoNLL-2009 dataset confirm that the proposed model substantially outperforms most previous SRL systems for both English and Chinese languages.
2015 IEEE 15th International Conference on Environment and Electrical Engineering (EEEIC), 2015
ABSTRACT Demand of power is increasing worldwide and this increased demand has motivated the engi... more ABSTRACT Demand of power is increasing worldwide and this increased demand has motivated the engineers to look for alternative or renewable energy systems. Meanwhile, the data centers are also employing the solar energy systems to avoid the AC/DC power conversion losses due to the fact that an efficient solar system with DC/DC converter will help them to save up to 15% of the power in comparison to the conventional power supply system using non-renewable energy sources. The efficiency of the solar energy systems is greatly dependent on the DC/DC converters and hence, there is a need to study the comparison of efficiency of such converters which can be used in Switch Model Power Supplies (SMPS). This paper provides a comparative analysis between some of the DC/DC converters i.e. synchronous buck converter, isolated buck converter and non-isolated buck converter. Moreover, the important factors such as input voltage variation and the effects of open and closed loop systems of mentioned DC/DC converters are also explained.