Vaidotas Drungilas

Mr. Vaidotas Drungilas

Faculty of Informatics, Kaunas University of Technology, Kaunas, Kauno Miestas, Lithuania
orcid

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

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

Representative Publications

View All Publications
28 October 2018 in Computer Applications in Engineering Education
Verified
Journal Article
Closed access
PDF file
Mantas JurgelaitisMantas Jurgelaitis
Mantas Jurgelaitis
Lina ČeponienėLina Čeponienė
Lina Čeponienė
JČJonas Čeponis
Jonas Čeponis
Vaidotas DrungilasVaidotas Drungilas
Vaidotas Drungilas

23 January 2021 in Applied Sciences
Verified
Journal Article
Open access
PDF file
Vaidotas DrungilasVaidotas Drungilas
Vaidotas Drungilas
EVEvaldas Vaiciukynas
Evaldas Vaiciukynas
Mantas JurgelaitisMantas Jurgelaitis
Mantas Jurgelaitis
Rita ButkienėRita Butkienė
Rita Butkienė
Lina ČeponienėLina Čeponienė
Lina Čeponienė
Federated learning is a branch of machine learning where a shared model is created in a decentralized and privacy-preserving fashion, but existing approaches using blockchain are limited by tailored models. We consider the possibility to extend a set of supported models by introducing the oracle service and exploring the usability of blockchain-based architecture. The investigated architecture combines an oracle service with a Hyperledger Fabric chaincode. We compared two logistic regression implementations in Go language—a pure chaincode and an oracle service—at various data (2–32 k instances) and network (3–13 peers) sizes. Experiments were run to assess the performance of blockchain-based model inference using 2D synthetic and EEG eye state datasets for a supervised machine learning detection task. The benchmarking results showed that the impact on performance is acceptable with the median overhead of oracle service reaching 2–4%, depending on the dimensionality of the dataset. The overhead tends to diminish at large dataset sizes with the runtime depending on the network size linearly, where additional peers increased the runtime by 6.3 and 6.6 s for 2D and EEG datasets, respectively. Demonstrated negligible difference between implementations justifies the flexible choice of model in the blockchain-based federated learning and other machine learning applications.

30 December 2022 in Applied Sciences
Verified
Journal Article
Open access
PDF file
Mantas JurgelaitisMantas Jurgelaitis
Mantas Jurgelaitis
Lina ČeponienėLina Čeponienė
Lina Čeponienė
KBKarolis Butkus
Karolis Butkus
Rita ButkienėRita Butkienė
Rita Butkienė
Vaidotas DrungilasVaidotas Drungilas
Vaidotas Drungilas
Blockchain smart contracts can support the decentralisation of business processes, but due to smart contracts’ specifics, their development is a complicated process. Introducing model-driven development principles in smart contract development can facilitate requirement specification, design, and implementation activities. This paper presents a model-driven development method MDAsmartCD (Model-Driven Architecture-based Smart Contract Development) to alleviate smart contract development by supporting the complete MDA life cycle, covering the definition of Computation-Independent Model, Platform-Independent Model, and two instances of Platform-Specific Models. In MDAsmartCD, model transformations (model-to-model and model-to-text) are used to produce smart contract code in the Hyperledger Fabric platform Go and the Ethereum platform Solidity programming languages. The method application was demonstrated by implementing the smart contract for the hackathon solution and executing the generated Solidity and Go smart contracts in the workflow of issuing certificates for hackathon participants. During the execution of the workflow, both deployed smart contracts behaved identically and recorded analogous results in respective blockchain data storages. This demonstrated that the MDAsmartCD method enables the generation of compilable and executable smart contract code, ready for deployment on a blockchain platform.

1 January 2018 in Baltic Journal of Modern Computing
Verified
Journal Article
Closed access
Mantas JurgelaitisMantas Jurgelaitis
Mantas Jurgelaitis
Vaidotas DrungilasVaidotas Drungilas
Vaidotas Drungilas
Lina ČeponienėLina Čeponienė
Lina Čeponienė

22 July 2024 in Applied Sciences
Verified
Journal Article
Open access
PDF file
PJPranas Jaruševičius
Pranas Jaruševičius
LPLukas Paulauskas
Lukas Paulauskas
Vaidotas DrungilasVaidotas Drungilas
Vaidotas Drungilas
Mantas JurgelaitisMantas Jurgelaitis
Mantas Jurgelaitis
Tomas BlažauskasTomas Blažauskas
Tomas Blažauskas
Currently, the integration of immersive virtual reality applications into learning management systems is limited, which disadvantages kinesthetic learners, as the majority of learning material is mostly written or visual in nature. This research presents a method on how to transform existing conventional learning material into immersive virtual reality and integrate it into learning management systems, demonstrating it with a pilot study and an immersive virtual reality application prototype. The prototype application uses WebXR technology and provides integration into the Moodle learning management system using the Sharable Content Object Reference Model standard. The developed learning object was evaluated using the System Usability Scale, Web-based Learning Tool evaluation scale and the igroup Presence Questionnaire. The results of the pilot study revealed that the learning object performed above average in all tested fields and is preferable to established non-IVR solutions.

10 June 2023 in Applied Sciences
Verified
Journal Article
Open access
PDF file
Vaidotas DrungilasVaidotas Drungilas
Vaidotas Drungilas
EVEvaldas Vaiciukynas
Evaldas Vaiciukynas
LALinas Ablonskis
Linas Ablonskis
Lina ČeponienėLina Čeponienė
Lina Čeponienė
This study introduces a novel performance-based weighting scheme for ensemble learning using the Shapley value. The weighting uses the reciprocal of binary cross-entropy as a base learner’s performance metric and estimates its Shapley value to measure the overall contribution of a learner to an equally weighted ensemble of various sizes. Two variants of this strategy were empirically compared with a single monolith model and other static weighting strategies using two large banking-related datasets. A variant that discards learners with a negative Shapley value was ranked as first or at least second when constructing homogeneous ensembles, whereas for heterogeneous ensembles this strategy resulted in a better or at least similar detection performance to other weighting strategies tested. The main limitation being the computational complexity of Shapley calculations, the explored weighting strategy could be considered as a generalization of performance-based weighting.