Nicolai Petkov
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Nicolai Petkov has a total of 62 co-authors. Here are the top co-authors by number of shared publications:
- George Azzopardi
University of Groningen
12 shared publications - Javier Gonzalez-Jimenez
Universidad de Málaga
6 shared publications - Nicola Strisciuglio
University of Twente
6 shared publications - Chenyu Shi
University of Groningen
4 shared publications - David Fernandez-Chaves
University of Malaga
4 shared publications - Estefania Talavera
University of Twente
4 shared publications
Latest Publications
- Top Cited PapersOpen Access
- 6 January 2026
- Published by Springer Nature in Financial Innovation
- Vol. 12 (1) , 14
Advances in forecasting realized volatility: a review of methodologies
Over the past decade, volatility modeling has gained increasing importance in quantitative finance, significantly influencing risk management, investment strategies, and policymaking. Traditionally, academic research has emphasized linear models for forecasting realized volatility. However, advances in computational power have enabled the development of more sophisticated approaches, including machine and deep learning models. Despite these advancements, no comprehensive survey currently compares traditional linear methods with newer models in the context of realized volatility forecasting. This survey addresses that gap by analyzing all models used to forecast realized volatility in academic literature from 2000 to the first half of 2024. It highlights key academic contributions and examines the empirical characteristics of realized volatility. Our findings indicate that the top-performing hybrid convolutional neural network and long short-term memory model surpasses other models in forecasting accuracy. These results highlight the ability of such models to capture a broader range of factors driving realized volatility, affirming their valuable role in quantitative modeling.
- 6 January 2026
- Published by Springer Nature in Financial Innovation
- Vol. 12 (1) , 14
Top Cited PapersOpen AccessAdvances in forecasting realized volatility: a review of methodologies
Over the past decade, volatility modeling has gained increasing importance in quantitative finance, significantly influencing risk management, investment strategies, and policymaking. Traditionally, academic research has emphasized linear models for forecasting realized volatility. However, advances in computational power have enabled the development of more sophisticated approaches, including machine and deep learning models. Despite these advancements, no comprehensive survey currently compares traditional linear methods with newer models in the context of realized volatility forecasting. This survey addresses that gap by analyzing all models used to forecast realized volatility in academic literature from 2000 to the first half of 2024. It highlights key academic contributions and examines the empirical characteristics of realized volatility. Our findings indicate that the top-performing hybrid convolutional neural network and long short-term memory model surpasses other models in forecasting accuracy. These results highlight the ability of such models to capture a broader range of factors driving realized volatility, affirming their valuable role in quantitative modeling.
- Open Access
- 1 January 2026
- Published by Elsevier in SSRN Electronic Journal
Advances in Forecasting Realized Volatility: A Review of Methodologies
Over the past decade, volatility modeling has become increasingly important in quantitative finance, significantly influencing risk management, investment decisions and policy making. Traditionally, academic research has focused on linear models for forecasting realized volatility. However, advances in computational power have opened the way for more sophisticated methods, including machine learning (ML) and deep learning (DL) models. However, currently, no survey compares traditional linear approaches with these newer ML and DL models in the context of realized volatility forecasting. This survey addresses this gap by analyzing all models used to forecast realized volatility in academic literature from 2000 to H1 2024. It highlights key academic articles and examines the empirical characteristics of realized volatility. Our findings show that the best performing model-a combination of CNN and LSTM networks-outperforms both linear and other ML models in forecasting accuracy. This highlights that DL models can capture a wider range of factors driving realized volatility, highlighting their valuable contribution to quantitative modeling.
- 1 January 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessAdvances in Forecasting Realized Volatility: A Review of Methodologies
Over the past decade, volatility modeling has become increasingly important in quantitative finance, significantly influencing risk management, investment decisions and policy making. Traditionally, academic research has focused on linear models for forecasting realized volatility. However, advances in computational power have opened the way for more sophisticated methods, including machine learning (ML) and deep learning (DL) models. However, currently, no survey compares traditional linear approaches with these newer ML and DL models in the context of realized volatility forecasting. This survey addresses this gap by analyzing all models used to forecast realized volatility in academic literature from 2000 to H1 2024. It highlights key academic articles and examines the empirical characteristics of realized volatility. Our findings show that the best performing model-a combination of CNN and LSTM networks-outperforms both linear and other ML models in forecasting accuracy. This highlights that DL models can capture a wider range of factors driving realized volatility, highlighting their valuable contribution to quantitative modeling.
- Open Access
Data-efficient Large Scale Place Recognition with Graded Similarity Supervision
Visual place recognition (VPR) is a fundamental task of computer vision for visual localization. Existing methods are trained using image pairs that either depict the same place or not. Such a binary indication does not consider continuous relations of similarity between images of the same place taken from different positions, determined by the continuous nature of camera pose. The binary similarity induces a noisy supervision signal into the training of VPR methods, which stall in local minima and require expensive hard mining algorithms to guarantee convergence. Motivated by the fact that two images of the same place only partially share visual cues due to camera pose differences, we deploy an automatic re-annotation strategy to re-label VPR datasets. We compute graded similarity labels for image pairs based on available localization metadata. Furthermore, we propose a new Generalized Contrastive Loss (GCL) that uses graded similarity labels for training contrastive networks. We demonstrate that the use of the new labels and GCL allow to dispense from hard-pair mining, and to train image descriptors that perform better in VPR by nearest neighbor search, obtaining superior or comparable results than methods that require expensive hard-pair mining and re-ranking techniques. Code and models available at: https://github.com/marialeyvallina/generalized_contrastive_loss
Open AccessData-efficient Large Scale Place Recognition with Graded Similarity Supervision
Visual place recognition (VPR) is a fundamental task of computer vision for visual localization. Existing methods are trained using image pairs that either depict the same place or not. Such a binary indication does not consider continuous relations of similarity between images of the same place taken from different positions, determined by the continuous nature of camera pose. The binary similarity induces a noisy supervision signal into the training of VPR methods, which stall in local minima and require expensive hard mining algorithms to guarantee convergence. Motivated by the fact that two images of the same place only partially share visual cues due to camera pose differences, we deploy an automatic re-annotation strategy to re-label VPR datasets. We compute graded similarity labels for image pairs based on available localization metadata. Furthermore, we propose a new Generalized Contrastive Loss (GCL) that uses graded similarity labels for training contrastive networks. We demonstrate that the use of the new labels and GCL allow to dispense from hard-pair mining, and to train image descriptors that perform better in VPR by nearest neighbor search, obtaining superior or comparable results than methods that require expensive hard-pair mining and re-ranking techniques. Code and models available at: https://github.com/marialeyvallina/generalized_contrastive_loss
- Open Access
- 5 December 2022
- Published by Elsevier in Preventive Veterinary Medicine
- Vol. 210, 105812
Pre-insemination prediction of dystocia in dairy cattle
- 5 December 2022
- Published by Elsevier in Preventive Veterinary Medicine
- Vol. 210, 105812
Open AccessPre-insemination prediction of dystocia in dairy cattle
- Open Access
LTC-Mapping, Enhancing Long-Term Consistency of Object-Oriented Semantic Maps in Robotics
This paper proposes LTC-Mapping, a method for building object-oriented semantic maps that remain consistent in the long-term operation of mobile robots. Among the different challenges that compromise this aim, LTC-Mapping focuses on two of the more relevant ones: preventing duplicate instances of objects (instance duplication) and handling dynamic scenes. The former refers to creating multiple instances of the same physical object in the map, usually as a consequence of partial views or occlusions. The latter deals with the typical assumption made by object-oriented mapping methods that the world is static, resulting in outdated representations when the objects change their positions. To face these issues, we model the detected objects with 3D bounding boxes, and analyze the visibility of their vertices to detect occlusions and partial views. Besides this geometric modeling, the boxes are augmented with semantic information regarding the categories of the objects they represent. Both the geometric entities (bounding boxes) and their semantic content are propagated over time through data association and a fusion technique. In addition, in order to keep the map curated, the non-detection of objects in the areas where they should appear is also considered, proposing a mechanism that removes them from the map once there is evidence that they have been moved (i.e., multiple non-detections occur). To validate our proposal, a number of experiments have been carried out using the Robot@VirtualHome ecosystem, comparing its performance with a state-of-the-art alternative. The results report a superior performance of LTC-Mapping when modeling both geometric and semantic information of objects, and also support its online execution.
Open AccessLTC-Mapping, Enhancing Long-Term Consistency of Object-Oriented Semantic Maps in Robotics
This paper proposes LTC-Mapping, a method for building object-oriented semantic maps that remain consistent in the long-term operation of mobile robots. Among the different challenges that compromise this aim, LTC-Mapping focuses on two of the more relevant ones: preventing duplicate instances of objects (instance duplication) and handling dynamic scenes. The former refers to creating multiple instances of the same physical object in the map, usually as a consequence of partial views or occlusions. The latter deals with the typical assumption made by object-oriented mapping methods that the world is static, resulting in outdated representations when the objects change their positions. To face these issues, we model the detected objects with 3D bounding boxes, and analyze the visibility of their vertices to detect occlusions and partial views. Besides this geometric modeling, the boxes are augmented with semantic information regarding the categories of the objects they represent. Both the geometric entities (bounding boxes) and their semantic content are propagated over time through data association and a fusion technique. In addition, in order to keep the map curated, the non-detection of objects in the areas where they should appear is also considered, proposing a mechanism that removes them from the map once there is evidence that they have been moved (i.e., multiple non-detections occur). To validate our proposal, a number of experiments have been carried out using the Robot@VirtualHome ecosystem, comparing its performance with a state-of-the-art alternative. The results report a superior performance of LTC-Mapping when modeling both geometric and semantic information of objects, and also support its online execution.
- Open Access
- 5 July 2022
- Published by Elsevier in Expert Systems with Applications
- Vol. 208
Robot@VirtualHome, an ecosystem of virtual environments and tools for realistic indoor robotic simulation
- 5 July 2022
- Published by Elsevier in Expert Systems with Applications
- Vol. 208
Open AccessRobot@VirtualHome, an ecosystem of virtual environments and tools for realistic indoor robotic simulation
- Open Access
- 1 January 2022
- Published by Elsevier in SSRN Electronic Journal
Pre-Insemination Prediction of Dystocia in Dairy Cattle
Dystocia or difficult calving in cattle is detrimental to health of the afflicted cows and has a negative economic impact on the dairy industry. The goal of this study was to create a data-driven tool for predicting the calving difficulty of non-heifer cows using input variables that are known prior to the moment of insemination. Compared to past studies, we excluded input variables that can only be known during or after insemination, such as birth weight and gestation length. This makes the model suitable for informing mating decisions that could reduce the incidence of difficult calvings or mitigate their consequences. We used a dataset consisting of 131,527 calving records of Dutch cattle, from which we derived a total of 274 phenotypic features and estimated breeding values. The distribution of classes in the dataset was 96.7\% normal calvings, and 3.3% difficult calvings. We used a gradient boosted trees (XGBoost) as the learning model and a bagging ensemble approach to deal with the extreme class imbalance. The model achieved an average area under the ROC curve of 0.73 on unseen test data. Using feature importance analysis, we identified a number of features that have a high discriminatory value for calving difficulty, including maternal and paternal breeding values, and past phenotypic measurements of the cow.
- 1 January 2022
- Published by Elsevier in SSRN Electronic Journal
Open AccessPre-Insemination Prediction of Dystocia in Dairy Cattle
Dystocia or difficult calving in cattle is detrimental to health of the afflicted cows and has a negative economic impact on the dairy industry. The goal of this study was to create a data-driven tool for predicting the calving difficulty of non-heifer cows using input variables that are known prior to the moment of insemination. Compared to past studies, we excluded input variables that can only be known during or after insemination, such as birth weight and gestation length. This makes the model suitable for informing mating decisions that could reduce the incidence of difficult calvings or mitigate their consequences. We used a dataset consisting of 131,527 calving records of Dutch cattle, from which we derived a total of 274 phenotypic features and estimated breeding values. The distribution of classes in the dataset was 96.7\% normal calvings, and 3.3% difficult calvings. We used a gradient boosted trees (XGBoost) as the learning model and a bagging ensemble approach to deal with the extreme class imbalance. The model achieved an average area under the ROC curve of 0.73 on unseen test data. Using feature importance analysis, we identified a number of features that have a high discriminatory value for calving difficulty, including maternal and paternal breeding values, and past phenotypic measurements of the cow.
- 31 October 2021
- Published by Springer Nature
- p. 186-196
Exploiting Spatio-Temporal Coherence for Video Object Detection in Robotics
- 31 October 2021
- Published by Springer Nature
- p. 186-196
Exploiting Spatio-Temporal Coherence for Video Object Detection in Robotics
- Open Access
- 11 December 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 8, 225475-225489
Virtual Reality for Pain Management in Cancer: A Comprehensive Review
Virtual Reality is a computer-simulated 3-Dimensional technology in which the user interacts via different senses: visual, auditory, tactile, and/or olfactory. In the past decades, it has been argued that Virtual Reality as a technique could be applied in the clinical environment to successfully manage pain. This article provides a systematic review of research on Virtual Reality and pain management for patients who are suffering from cancer. More specifically, this article focuses on all types of Virtual Reality technologies (Non-Immersive, Semi-Immersive, Fully-Immersive) which has been developed and released to manage the pain which evokes from the treatment of cancer. An exhaustive search identified 23 relevant studies from 2010 to 2020. Overall, the identified studies indicated that Virtual Reality can improve the experience of pain for patients who are suffering from cancer. It was also found that, if Virtual Reality is appropriately designed, the pain which is arising from cancer treatments can be reduced. Even though some positive outcomes have been reported, overall, the results are inconclusive and studies that examine specifically the treatment of pain in cancer patients are limited. Further research needs to be conducted, to articulate clearly, under what circumstances Virtual Reality is an effective tool for cancer patients, and under what factors Virtual Reality can be the solution to the pain patients are experiencing.
- 11 December 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 8, 225475-225489
Open AccessVirtual Reality for Pain Management in Cancer: A Comprehensive Review
Virtual Reality is a computer-simulated 3-Dimensional technology in which the user interacts via different senses: visual, auditory, tactile, and/or olfactory. In the past decades, it has been argued that Virtual Reality as a technique could be applied in the clinical environment to successfully manage pain. This article provides a systematic review of research on Virtual Reality and pain management for patients who are suffering from cancer. More specifically, this article focuses on all types of Virtual Reality technologies (Non-Immersive, Semi-Immersive, Fully-Immersive) which has been developed and released to manage the pain which evokes from the treatment of cancer. An exhaustive search identified 23 relevant studies from 2010 to 2020. Overall, the identified studies indicated that Virtual Reality can improve the experience of pain for patients who are suffering from cancer. It was also found that, if Virtual Reality is appropriately designed, the pain which is arising from cancer treatments can be reduced. Even though some positive outcomes have been reported, overall, the results are inconclusive and studies that examine specifically the treatment of pain in cancer patients are limited. Further research needs to be conducted, to articulate clearly, under what circumstances Virtual Reality is an effective tool for cancer patients, and under what factors Virtual Reality can be the solution to the pain patients are experiencing.