Nicolae CONSTANTIN

Prof. Nicolae CONSTANTIN

Research Institute for Artificial Inteligence "Mihai Draganescu", Romanian Academy
orcid

Research Focus & Metrics

Academic Career Length

17Years (2010-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
3 August 2021 in Computers
Verified
Journal Article
Open access
PDF file
VEVictoria Eyharabide
Victoria Eyharabide
Imad Eddine Ibrahim BEKKOUCHImad Eddine Ibrahim BEKKOUCH
Imad Eddine Ibrahim BEKKOUCH
NCNicolae CONSTANTIN
Nicolae CONSTANTIN
Convolutional neural networks raised the bar for machine learning and artificial intelligence applications, mainly due to the abundance of data and computations. However, there is not always enough data for training, especially when it comes to historical collections of cultural heritage where the original artworks have been destroyed or damaged over time. Transfer Learning and domain adaptation techniques are possible solutions to tackle the issue of data scarcity. This article presents a new method for domain adaptation based on Knowledge graph embeddings. Knowledge Graph embedding forms a projection of a knowledge graph into a lower-dimensional where entities and relations are represented into continuous vector spaces. Our method incorporates these semantic vector spaces as a key ingredient to guide the domain adaptation process. We combined knowledge graph embeddings with visual embeddings from the images and trained a neural network with the combined embeddings as anchors using an extension of Fisher’s linear discriminant. We evaluated our approach on two cultural heritage datasets of images containing medieval and renaissance musical instruments. The experimental results showed a significant increase in the baselines and state-of-the-art performance compared with other domain adaptation methods.

15 March 2021 in IEEE Access
Verified
Journal Article
Open access
Imad Eddine Ibrahim BEKKOUCHImad Eddine Ibrahim BEKKOUCH
Imad Eddine Ibrahim BEKKOUCH
NCNicolae CONSTANTIN
Nicolae CONSTANTIN
AKAdil Khan
Adil Khan
SKS. M. Ahsan Kazmi
S. M. Ahsan Kazmi
AKAsad Masood Khattak
Asad Masood Khattak
BIBulat Ibragimov
Bulat Ibragimov
The biggest fear when deploying machine learning models to the real world is their ability to handle the new data. This problem is significant especially in medicine, where models trained on rich high-quality data extracted from large hospitals do not scale to small regional hospitals. One of the clinical challenges addressed in this work is magnetic resonance image generalization for improved visualization and diagnosis of hip abnormalities such as femoroacetabular impingement and dysplasia. Domain Generalization (DG) is a field in machine learning that tries to solve the model's dependency on the training data by leveraging many related but different data sources. We present a new method for DG that is both efficient and fast, unlike the most current state of art methods, which add a substantial computational burden making it hard to fine-tune. Our model trains an autoencoder setting on top of the classifier, but the encoder is trained on the adversarial reconstruction loss forcing it to forget style information while extracting features useful for classification. Our approach aims to force the encoder to generate domain-invariant representations that are still category informative by pushing it in both directions. Our method has proven universal and was validated on four different benchmarks for domain generalization, outperforming state of the art on RMNIST, VLCS and IXMAS with a 0.70% increase in accuracy and providing comparable results on PACS with a 0.02% difference. Our method was also evaluated for unsupervised domain adaptation and has shown to be quite an effective method against over-fitting.

15 April 2022 in Computers
Verified
Journal Article
Open access
PDF file
NCNicolae CONSTANTIN
Nicolae CONSTANTIN
RYRohan Kumar Yadav
Rohan Kumar Yadav
DTDan Ioan Tufis
Dan Ioan Tufis
Large-scale pre-trained language representation and its promising performance in various downstream applications have become an area of interest in the field of natural language processing (NLP). There has been huge interest in further increasing the model’s size in order to outperform the best previously obtained performances. However, at some point, increasing the model’s parameters may lead to reaching its saturation point due to the limited capacity of GPU/TPU. In addition to this, such models are mostly available in English or a shared multilingual structure. Hence, in this paper, we propose a lite BERT trained on a large corpus solely in the Romanian language, which we called “A Lite Romanian BERT (ALR-BERT)”. Based on comprehensive empirical results, ALR-BERT produces models that scale far better than the original Romanian BERT. Alongside presenting the performance on downstream tasks, we detail the analysis of the training process and its parameters. We also intend to distribute our code and model as an open source together with the downstream task.

22 April 2022 in Algorithms
Verified
Journal Article
Open access
PDF file
RYRohan Kumar Yadav
Rohan Kumar Yadav
NCNicolae CONSTANTIN
Nicolae CONSTANTIN
Explainability is one of the key factors in Natural Language Processing (NLP) specially for legal documents, medical diagnosis, and clinical text. Attention mechanism has been a popular choice for such explainability recently by estimating the relative importance of input units. Recent research has revealed, however, that such processes tend to misidentify irrelevant input units when explaining them. This is due to the fact that language representation layers are initialized by pre-trained word embedding that is not context-dependent. Such a lack of context-dependent knowledge in the initial layer makes it difficult for the model to concentrate on the important aspects of input. Usually, this does not impact the performance of the model, but the explainability differs from human understanding. Hence, in this paper, we propose an ensemble method to use logic-based information from the Tsetlin Machine to embed it into the initial representation layer in the neural network to enhance the model in terms of explainability. We obtain the global clause score for each word in the vocabulary and feed it into the neural network layer as context-dependent information. Our experiments show that the ensemble method enhances the explainability of the attention layer without sacrificing any performance of the model and even outperforming in some datasets.

9 February 2023 in International Journal of Computers Communications & Control
Verified
Journal Article
Open access
PDF file
NCNicolae CONSTANTIN
Nicolae CONSTANTIN
RYRohan Kumar Yadav
Rohan Kumar Yadav
DTDan Ioan Tufis
Dan Ioan Tufis
Natural language processing (NLP) has become a vital requirement in a wide range of applications, including machine translation, information retrieval, and text classification. The development and evaluation of NLP models for various languages have received significant attention in recent years, but there has been relatively little work done on comparing the performance of different language models on Romanian data. In particular, the introduction and evaluation of various Romanian language models with multilingual models have barely been comparatively studied. In this paper, we address this gap by evaluating eight NLP models on two Romanian datasets, XQuAD and RoITD. Our experiments and results show that bert-base-multilingual-cased and bertbase- multilingual-uncased, perform best on both XQuAD and RoITD tasks, while RoBERT-small model and DistilBERT models perform the worst. We also discuss the implications of our findings and outline directions for future work in this area.

1 January 2010 in The International Journal of Acoustics and Vibration
Verified
Journal Article
Open access
NCNicolae CONSTANTIN
Nicolae CONSTANTIN
SSStefan Sorohan
Stefan Sorohan
MGMircea Gavan
Mircea Gavan
VAViorel Anghel
Viorel Anghel