Jivitesh Sharma

• Alam, Azimil Gani; Bartonova, Alena; Sharma, Jivitesh; Fredriksen, Mirjam; Høiskar, Britt Ann Kåstad & Mathisen, Hans Martin [Vis alle 10 forfattere av denne artikkelen] (2026). Machine Learning Prediction of Building-Related Symptoms Based-on Indoor Environment Complaints: Study Case in a Norwegian School. I Zilio, Claudio; Busato, Filippo; Mazzarella, Livio & Noro, Marco (Red.), Proceedings of the 15th REHVA HVAC World Congress - CLIMA 2025. Springer Nature Switzerland. ISSN 9783032068101. s. 379–389. doi: 10.1007/978-3-032-06810-1_39. Fulltekst i vitenarkiv
• Sharma, Jivitesh; Vallejo, Islen; Ødegård, Rune Åvar; Le, Truong Thanh; Taherkordi, Amirhosein & Eliassen, Frank (2025). Physics-Informed Deep Learning for Wind Downscaling over Oslo. Proceedings - International Conference on Tools with Artificial Intelligence (ICTAI). ISSN 1082-3409. 37, s. 270–276. doi: 10.1109/ICTAI66417.2025.00042. Fulltekst i vitenarkiv Vis sammendrag

  Running a numerical weather model such as WRF at kilometre or sub-kilometre grid spacing over a regional domain is computationally expensive. We present physics-informed deeplearning models that ingest a single 9km WRF wind field and simultaneously predict two finer-scale wind fields at 3 km and 1 km resolution via dual decoder heads. Four representative architectures are benchmarked-Deep Residual U-Net (DeepRU), DEVINE, a bespoke 3-D Transformer, and a Fourier Neural Operator (FNO)-each trained with divergence-free, vorticity, and Navier-Stokes residual constraints plus Charbonnier and gradient perceptual losses. We train and validate our models on the city of Oslo for the year 2018. DeepRU achieves R2=0.94 (RMSE =0.050) at 3km and R2=0.89(RMSE=0.065) at 1 km. DEVINE, Transformer 3-D, and FNO yield 3 km scores of 0.91−0.93, with 1km scores lower by 0.02−0.08, illustrating the increased difficulty of finer-scale reconstruction. Physicsinformed losses improve all models compared to MSE-only baselines, and the residual architecture (DeepRU) remains most effective for this dual-scale task.

• Alam, Azimil Gani; Bartonova, Alena; Høiskar, Britt Ann Kåstad; Fredriksen, Mirjam; Sharma, Jivitesh & Mathisen, Hans Martin [Vis alle 11 forfattere av denne artikkelen] (2025). Building-related symptoms in school environment: Predictability using machine learning approach. Indoor Environments. 2(4). doi: 10.1016/j.indenv.2025.100140. Fulltekst i vitenarkiv Vis sammendrag

  Building-related symptoms (BRS) are commonly experienced by students in schools and are potentially affecting academic performance and health. Even though indoor environment quality (IEQ) measurements indicated fair conditions, students still perceived discomfort that led to symptoms, highlighting the necessity of collecting user-feedback about IEQ-complaints. This study aimed to predict and understand the prevalence of BRS (headache, tiredness, cough, dry eyes-hands) experienced by students in classrooms using machine-learning (ML) approach based on measurement data, building factors, and prevalence of IEQ-complaints. We collected measurement data (from indoor and outdoor climate), building factors, and user-feedback by students via online-platform across three sampled classrooms each campaign during three consecutive school semesters. Significant input variables for ML were pre-selected using statistical tests. ML models were evaluated based on accuracy metrics and SHAP analysis for input interpretation. Models using measurement data alone performed poorly (testing R² <50 %) to predict prevalence of BRS, whereas adding building factors and prevalence of IEQ-complaints increased accuracy (R² up to 95 %) of prediction of BRS with lower RMSE. In addition, interpretation from SHAP analysis showed IEQ-complaints especially related with indoor air quality (e.g., heavy air, dust & dirt, and dry air) as significant contributors for predicting prevalence of BRS. We conclude that the framework of combining objective measurements with occupant-reported complaints can be reliable, interpretable predictions of symptom prevalence. This study is limited by single-school setting, health confounders, and symptoms verification. Future research may contribute to exploring wider set of input variables, applicability, and variation of complaints preference.

• Sharma, Jivitesh; Vallejo, Islen & Ødegård, Rune Åvar (2025). AirQ-ResUNet: A Residual U-Net Based Deep Learning Surrogate for High-Resolution PM2.5 Prediction in Urban Environments. I Kahraman, Cengiz; Cebi, Selcuk; Oztaysi, Basar; Onar, Sezi Cevik; Tolga, Cagrı; Sari, Irem Ucal & Otay, Irem (Red.), Intelligent and Fuzzy Systems: Artificial Intelligence in Human-Centric, Resilient and Sustainable Industries, Proceedings of the INFUS 2025 Conference, Volume 2. Springer. ISSN 9783031979927. s. 21–29. doi: 10.1007/978-3-031-97992-7_3. Fulltekst i vitenarkiv
• Belaid, Mohamed-Bachir; Sharma, Jivitesh; Lei, Jiao; Granmo, Ole-Christoffer; Andersen, Per-Arne & Yazidi, Anis (2025). Generalized Convergence Analysis of Tsetlin Automaton Based Algorithms: A Probabilistic Approach to Concept Learning. I Walsh, Toby; Shah, Julie & Kolter, Zico (Red.), Proceedings of the Thirty-Ninth AAAI Conference on Artificial Intelligence. AAAI Press. ISSN 9781577358978. s. 15524–15531. doi: 10.1609/aaai.v39i15.33704. Fulltekst i vitenarkiv
• Alam, Azimil Gani; Bartonova, Alena; Høiskar, Britt Ann Kåstad; Fredriksen, Mirjam; Sharma, Jivitesh & Mathisen, Hans Martin [Vis alle 11 forfattere av denne artikkelen] (2025). Predicting the student's perceptions of multi-domain environmental factors in a Norwegian school building: Machine learning approach. Building and Environment. ISSN 0360-1323. 280. doi: 10.1016/j.buildenv.2025.113144. Fulltekst i vitenarkiv Vis sammendrag

  Poor Indoor Environmental Quality (IEQ) in schools significantly impacts students’ well-being, learning capabilities, and health. Perceived dissatisfaction rates (PD%) among students often remain high, even when indoor environmental variables appear well-controlled. This study aims to predict perceived dissatisfaction rates (PD%) across multi-domain environmental factors—thermal, acoustic, visual, and indoor air quality (IAQ)—using machine learning (ML) models. The research integrates sensor-based environmental measurements, outdoor weather data, building parameters, and 1437 student survey responses collected from three classrooms in a Norwegian school across multiple seasons. Statistical tests were used to pre-select relevant input variables, followed by the development and evaluation of multiple ML algorithms. Among the tested ML models, Random Forest (RF) demonstrated the highest predictive accuracy for PD%, outperforming multi-linear regression (MLR) and decision trees (DT), with R² values up to 0.91 for overall IEQ dissatisfaction (PDIEQ%). SHAP analysis revealed key predictors: CO₂ levels, VOCs, humidity, temperature, solar radiation, and room window orientation. IAQ, thermal comfort, and acoustic environment were the most influential factors affecting students' perceived well-being. Despite limitations as implementation in building level scale, the study demonstrates the feasibility of deploying predictive ML models under real-world constraints for improving IEQ monitoring system. The findings support practical strategies for adaptive indoor environmental management, particularly in educational settings, and provide a replicable framework for future research. Future research can expand to other climates, buildings, measurements, occupant levels, and ML training optimization.

• Sharma, Jivitesh; Jetschny, Stefan; Kapun, Martin & Belaid, Mohamed-Bachir (2024). Deep Learning-Enhanced Gap Filling in Drosophila Melanogaster Genomic Data, ICMLA 2024- 2024 International Conference on Machine Learning and Applications. IEEE (Institute of Electrical and Electronics Engineers). ISSN 9798350374889. s. 1447–1453. doi: 10.1109/ICMLA61862.2024.00224. Fulltekst i vitenarkiv
• Sharma, Jivitesh; Jetschny, Stefan; Maza, Miquel S.; Guardia, Nuria B.; Peris, Eulàlia & Esteve, Jaume F. (2024). Deep Neural Networks for Comprehensive Environmental Noise Estimation in European Cities, ICMLA 2024- 2024 International Conference on Machine Learning and Applications. IEEE (Institute of Electrical and Electronics Engineers). ISSN 9798350374889. s. 71–78. doi: 10.1109/ICMLA61862.2024.00016. Fulltekst i vitenarkiv
• Bhattarai, Bimal; Granmo, Ole-Christoffer; Jiao, Lei; Yadav, Rohan Kumar & Sharma, Jivitesh (2024). Tsetlin Machine Embedding: Representing Words Using Logical Expressions, Findings of the Association for Computational Linguistics: EACL 2024. Association for Computational Linguistics (ACL). ISSN 9798891760936. s. 1512–1520. Fulltekst i vitenarkiv
• Sharma, Jivitesh; Yadav, Rohan Kumar; Granmo, Ole-Christoffer & Lei, Jiao (2023). Drop Clause: Enhancing Performance, Robustness and Pattern Recognition Capabilities of the Tsetlin Machine. I Chen, Yiling & Neville, Jennifer (Red.), Proceedings of the Thirty-Seventh AAAI Conference on Artificial Intelligence. AAAI Press. ISSN 9781577358800. s. 13547–13555. doi: 10.1609/aaai.v37i11.26588. Fulltekst i vitenarkiv
• Borgersen, Karl Audun Kagnes; Goodwin, Morten; Sharma, Jivitesh; Aasmoe, Tobias; Leonhardsen, Mari & Rørvik, Gro Herredsvela (2023). CorrEmbed: Evaluating Pre-trained Model Image Similarity Efficacy with a Novel Metric. I Bramer, Max & Stahl, Frederic (Red.), Artificial Intelligence XL: 43rd SGAI International Conference on Artificial Intelligence, AI 2023. Springer. ISSN 9783031479939. s. 89–102. doi: 10.1007/978-3-031-47994-6_7. Fulltekst i vitenarkiv
• Przybysz, Emilia Katarzyna; Bhattarai, Bimal; Persia, Cosimo Damiano; Ozaki, Ana; Granmo, Ole-Christoffer & Sharma, Jivitesh (2023). Verifying Properties of Tsetlin Machines, 2023 International Symposium on the Tsetlin Machine (ISTM). IEEE (Institute of Electrical and Electronics Engineers). ISSN 9798350344776. doi: 10.1109/ISTM58889.2023.10454997. Fulltekst i vitenarkiv
• Abeyrathna, Kuruge Darshana; Abouzeid, Ahmed Abdulrahem Othman; Bhattarai, Bimal; Giri, Charul; Glimsdal, Sondre & Granmo, Ole-Christoffer [Vis alle 11 forfattere av denne artikkelen] (2023). Building Concise Logical Patterns by Constraining Tsetlin Machine Clause Size. I Elkind, Edith (Red.), Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence. AAAI Press. ISSN 9781956792034. s. 3395–3403. doi: 10.24963/ijcai.2023/378. Fulltekst i vitenarkiv Vis sammendrag

  Tsetlin machine (TM) is a logic-based machine learning approach with the crucial advantages of being transparent and hardware-friendly. While TMs match or surpass deep learning accuracy for an increasing number of applications, large clause pools tend to produce clauses with many literals (long clauses). As such, they become less interpretable. Further, longer clauses increase the switching activity of the clause logic in hardware, consuming more power. This paper introduces a novel variant of TM learning – Clause Size Constrained TMs (CSC-TMs) – where one can set a soft constraint on the clause size. As soon as a clause includes more literals than the constraint allows, it starts expelling literals. Accordingly, oversized clauses only appear transiently. To evaluate CSC-TM, we conduct classifcation, clustering, and regression experiments on tabular data, natural language text, images, and board games. Our results show that CSC-TM maintains accuracy with up to 80 times fewer literals. Indeed, the accuracy increases with shorter clauses for TREC, IMDb, and BBC Sports. After the accuracy peaks, it drops gracefully as the clause size approaches a single literal. We fnally analyze CSC-TM power consumption and derive new convergence properties.

• Berge, Geir Thore; Granmo, Ole-Christoffer; Tveit, Tor Oddbjørn; Ruthjersen, Anna Linda & Sharma, Jivitesh (2023). Combining unsupervised, supervised and rule-based learning: the case of detecting patient allergies in electronic health records. BMC Medical Informatics and Decision Making. 23. doi: 10.1186/s12911-023-02271-8. Fulltekst i vitenarkiv Vis sammendrag

  Background Data mining of electronic health records (EHRs) has a huge potential for improving clinical decision support and to help healthcare deliver precision medicine. Unfortunately, the rule-based and machine learning-based approaches used for natural language processing (NLP) in healthcare today all struggle with various shortcomings related to performance, efciency, or transparency. Methods In this paper, we address these issues by presenting a novel method for NLP that implements unsupervised learning of word embeddings, semi-supervised learning for simplifed and accelerated clinical vocabulary and concept building, and deterministic rules for fne-grained control of information extraction. The clinical language is automatically learnt, and vocabulary, concepts, and rules supporting a variety of NLP downstream tasks can further be built with only minimal manual feature engineering and tagging required from clinical experts. Together, these steps create an open processing pipeline that gradually refnes the data in a transparent way, which greatly improves the interpretable nature of our method. Data transformations are thus made transparent and predictions interpretable, which is imperative for healthcare. The combined method also has other advantages, like potentially being language independent, demanding few domain resources for maintenance, and able to cover misspellings, abbreviations, and acronyms. To test and evaluate the combined method, we have developed a clinical decision support system (CDSS) named Information System for Clinical Concept Searching (ICCS) that implements the method for clinical concept tagging, extraction, and classifcation. Results In empirical studies the method shows high performance (recall 92.6%, precision 88.8%, F-measure 90.7%), and has demonstrated its value to clinical practice. Here we employ a real-life EHR-derived dataset to evaluate the method’s performance on the task of classifcation (i.e., detecting patient allergies) against a range of common supervised learning algorithms. The combined method achieves state-of-the-art performance compared to the alternative methods we evaluate. We also perform a qualitative analysis of common word embedding methods on the task of word similarity to examine their potential for supporting automatic feature engineering for clinical NLP tasks. Conclusions Based on the promising results, we suggest more research should be aimed at exploiting the inherent synergies between unsupervised, supervised, and rule-based paradigms for clinical NLP.

• Berge, Geir Thore; Granmo, Ole-Christoffer; Tveit, Tor Oddbjørn; Munkvold, Bjørn Erik; Ruthjersen, Anna Linda & Sharma, Jivitesh (2023). Machine learning-driven clinical decision support system for concept-based searching: a field trial in a Norwegian hospital. BMC Medical Informatics and Decision Making. 23. doi: 10.1186/s12911-023-02101-x. Fulltekst i vitenarkiv Vis sammendrag

  Background Natural language processing (NLP) based clinical decision support systems (CDSSs) have demonstrated the ability to extract vital information from patient electronic health records (EHRs) to facilitate important decision support tasks. While obtaining accurate, medical domain interpretable results is crucial, it is demanding because realworld EHRs contain many inconsistencies and inaccuracies. Further, testing of such machine learning-based systems in clinical practice has received limited attention and are yet to be accepted by clinicians for regular use. Methods We present our results from the evaluation of an NLP-driven CDSS developed and implemented in a Norwegian Hospital. The system incorporates unsupervised and supervised machine learning combined with rulebased algorithms for clinical concept-based searching to identify and classify allergies of concern for anesthesia and intensive care. The system also implements a semi-supervised machine learning approach to automatically annotate medical concepts in the narrative. Results Evaluation of system adoption was performed by a mixed methods approach applying The Unifed Theory of Acceptance and Use of Technology (UTAUT) as a theoretical lens. Most of the respondents demonstrated a high degree of system acceptance and expressed a positive attitude towards the system in general and intention to use the system in the future. Increased detection of patient allergies, and thus improved quality of practice and patient safety during surgery or ICU stays, was perceived as the most important advantage of the system. Conclusions Our combined machine learning and rule-based approach benefts system performance, efciency, and interpretability. The results demonstrate that the proposed CDSS increases detection of patient allergies, and that the system received high-level acceptance by the clinicians using it. Useful recommendations for further system improvements and implementation initiatives are reducing the quantity of alarms, expansion of the system to include more clinical concepts, closer EHR system integration, and more workstations available at point of care.

• Borgersen, Karl Audun Kagnes; Goodwin, Morten & Sharma, Jivitesh (2023). A comparison between Tsetlin machines and deep neural networks in the context of recommendation systems. Proceedings of the Northern Lights Deep Learning Workshop. doi: 10.7557/18.6807. Fulltekst i vitenarkiv Vis sammendrag

  Recommendation Systems (RSs) are ubiquitous inmodern society and are one of the largest points ofinteraction between humans and AI. Modern RSsare often implemented using deep learning models,which are infamously difficult to interpret. Thisproblem is particularly exasperated in the contextof recommendation scenarios, as it erodes the user’strust in the RS. In contrast, the newly introducedTsetlin Machines (TM) possess some valuable prop-erties due to their inherent interpretability. TMsare still fairly young as a technology. As no RShas been developed for TMs before, it has becomenecessary to perform some preliminary research re-garding the practicality of such a system. In thispaper, we develop the first RS based on TMs toevaluate its practicality in this application domain.This paper compares the viability of TMs withother machine learning models prevalent in the fieldof RS. We train and investigate the performance ofthe TM compared with a vanilla feed-forward deeplearning model. These comparisons are based onmodel performance, interpretability/explainability,and scalability. Further, we provide some bench-mark performance comparisons to similar machinelearning solutions relevant to RSs.

• Giri, Charul; Sharma, Jivitesh & Goodwin, Morten (2022). Brain Tumour Segmentation on 3D MRI Using Attention V-Net. I Iliadis, Lazaros; Jayne, Chrisina; Tefas, Anastasios & Pimenidis, Elias (Red.), Engineering Applications of Neural Networks. EANN 2022. Communications in Computer and Information Science. Springer Nature. ISSN 9783031082238. s. 336–348. doi: 10.1007/978-3-031-08223-8_28. Fulltekst i vitenarkiv Vis sammendrag

  Brain tumour segmentation on 3D MRI imaging is one of the most critical deep learning applications. In this paper, for the segmentation of tumour sub-regions in brain MRI images, we study some popular architecture for medical imaging segmentation. We further, inspired by them, proposed an architecture that is an end-to-end trainable, fully convolutional neural network that uses attention block to learn localization of different features of the multiple sub-regions of a tumour. We also experiment with a combination of the weighted cross-entropy loss function and dice loss function on the model’s performance and the quality of the output segmented labels. The results of the evaluation of our model are received through BraTS’19 dataset challenge. The model can achieve a dice score of 0.80 for the whole tumour segmentation and dice scores of 0.639 and 0.536 for the other two sub-regions within the tumour on the validation dataset.

• Glimsdal, Sondre; Saha, Rupsa; Bhattarai, Bimal; Giri, Charul; Sharma, Jivitesh & Tunheim, Svein Anders [Vis alle 7 forfattere av denne artikkelen] (2022). Focused Negative Sampling for Increased Discriminative Power in Tsetlin Machines. I Shafik, Rishad (Red.), 2022 International Symposium on the Tsetlin Machine (ISTM 2022). IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781665471169. s. 73–80. doi: 10.1109/ISTM54910.2022.00021. Fulltekst i vitenarkiv Vis sammendrag

  Tsetlin Machines learn from input data by creating patterns in propositional logical, using the literals available in the data. These patterns vote for the classes in a classification task. Despite their simplistic premise, Tsetlin machine (TM)s have been performing at with other popular machine learning methods across various benchmarks. Not only accuracy, TMs also perform well in terms of energy efficiency and learning speed. The general TM scheme works best when there is sufficient discriminatory information available between two classes. In this paper, we explore the use of focused negative sampling (FNS) to discriminate between classes which are not easily distinguishable from each other. We carry out experiments across diverse classification tasks ranging over natural language processing, image processing, reinforcement learning to show that this approach forces the TM to arrive at patterns that can successfully tell apart two classes that are correlated. Further, we show that the proposed method achieves accuracy comparable to a vanilla Tsetlin Machine approach but in approximately 42% less number of epochs on average

• Sharma, Jivitesh; Granmo, Ole-Christoffer & Goodwin, Morten (2021). Emergency Analysis: Multitask Learning with Deep Convolutional Neural Networks for Fire Emergency Scene Parsing. Lecture Notes in Computer Science (LNCS). ISSN 0302-9743. Fulltekst i vitenarkiv
• Sharma, Jivitesh; Andersen, Per-Arne; Granmo, Ole-Christoffer & Goodwin, Morten (2021). Deep Q-Learning with Q-Matrix Transfer Learning for Novel Fire Evacuation Environment. IEEE Transactions on Systems, Man & Cybernetics. Systems. ISSN 2168-2216. 51(12), s. 7363–7381. doi: 10.1109/TSMC.2020.2967936. Fulltekst i vitenarkiv
• Sharma, Jivitesh; Granmo, Ole-Christoffer & Goodwin, Morten (2020). Environment Sound Classification using Multiple Feature Channels and Attention based Deep Convolutional Neural Network. Interspeech (USB). ISSN 2308-457X. 2020, s. 1186–1190. doi: 10.21437/Interspeech.2020-1303. Fulltekst i vitenarkiv Vis sammendrag

  In this paper, we propose a model for the Environment Sound Classification Task (ESC) that consists of multiple feature channels given as input to a Deep Convolutional Neural Network (CNN) with Attention mechanism. The novelty of the paper lies in using multiple feature channels consisting of Mel-Frequency Cepstral Coefficients (MFCC), Gammatone Frequency Cepstral Coefficients (GFCC), the Constant Q-transform CQT) and Chromagram. And, we employ a deeper CNN (DCNN) compared to previous models, consisting of spatially separable convolutions working on time and feature domain separately. Alongside, we use attention odules that perform channel and spatial attention together. We use the mix-up data augmentation technique to further boost performance. Our model is able to achieve state-of-the-art performance on three enchmark environment sound classification datasets, i.e. the UrbanSound8K (97.52%), ESC-10 (94.75%) and ESC-50 (87.45%).

• Sharma, Jivitesh; Granmo, Ole-Christoffer & Goodwin, Morten (2020). Emergency Detection with Environment Sound Using Deep Convolutional Neural Networks. I Yang, Xin-She; Sherratt, Simon; Dey, Nilanjan & Joshi, Amit (Red.), Proceedings of Fifth International Congress on Information and Communication Technology. Springer Nature. ISSN 9789811558580. s. 144–154. doi: 10.1007/978-981-15-5859-7_14. Fulltekst i vitenarkiv Vis sammendrag

  In this paper we propose a generic emergency detection system using only the sound produced in the environment. For this task, we employ multiple audio feature extraction techniques like the Mel-Frequency Cepstral Coefficients, Gammatone Frequency Cepstral Coefficients, Constant Q-transform and Chromagram. After feature extraction, a Deep Convolutional Neural Network (CNN) is used to classify an audio signal as a potential emergency situation or not. The entire model is based on our previous work that set the new state-of-the-art in the Environment Sound Classification (ESC) task1. We combine the benchmark ESC datasets: UrbanSound8K and ESC-50 (ESC-10 is a subset of ESC-50), and reduce the problem to a binary classification problem. This is done by aggregating sound classes such as sirens, fire crackling, glass breaking, gun shot as the emergency class and others as normal. Even though there are only two classes to distinguish, they are highly imbalanced. To overcome this difficulty we introduce class weights in calculating the loss while training the model. Our model is able to achieve 99.56% emergency detection accuracy.

• Sharma, Jivitesh; Granmo, Ole-Christoffer & Goodwin, Morten (2019). Environment Sound Classification using Multiple Feature Channels and Deep Convolutional Neural Networks. arXiv.org. ISSN 2331-8422. Fulltekst i vitenarkiv Vis sammendrag

  In this paper, we propose a model for the Environment Sound Classification Task (ESC) that consists of multiple feature channels given as input to a Deep Convolutional Neural Network (CNN). The novelty of the paper lies in using multiple feature channels consisting of Mel-Frequency Cepstral Coefficients (MFCC), Gammatone Frequency Cepstral Coefficients (GFCC), the Constant Q-transform (CQT) and Chromagram. Such multiple features have never been used before for signal or audio processing. Also, we employ a deeper CNN (DCNN) compared to previous models, consisting of 2D separable convolutions working on time and feature domain separately. The model also consists of max pooling layers that downsample time and feature domain separately. We use some data augmentation techniques to further boost performance. Our model is able to achieve state-of-the-art performance on all three benchmark environment sound classification datasets, i.e. the UrbanSound8K (97.35%), ESC-10 (95.75%) and ESC-50 (90.48%). To the best of our knowledge, this is the first time that a single environment sound classification model is able to achieve state-of-the-art results on all three datasets. For ESC-10 and ESC-50 datasets, the accuracy achieved by the proposed model is beyond human accuracy of 95.7% and 81.3% respectively.

• Matheussen, Bernt Viggo; Granmo, Ole-Christoffer & Sharma, Jivitesh (2019). Hydropower optimization using deep learning. Lecture Notes in Computer Science (LNCS). ISSN 0302-9743. 11606 LNAI, s. 110–122. doi: 10.1007/978-3-030-22999-3_11. Fulltekst i vitenarkiv
• Sharma, Jivitesh; Andersen, Per-Arne; Granmo, Ole-Christoffer & Goodwin, Morten (2019). Deep Q-Learning with Q-Matrix Transfer Learning for Novel Fire Evacuation Environment. arXiv.org. ISSN 2331-8422. doi: 10.1109/tsmc.2020.2967936. Fulltekst i vitenarkiv Vis sammendrag

  We focus on the important problem of emergency evacuation, which clearly could benefit from reinforcement learning that has been largely unaddressed. Emergency evacuation is a complex task which is difficult to solve with reinforcement learning, since an emergency situation is highly dynamic, with a lot of changing variables and complex constraints that makes it difficult to train on. In this paper, we propose the first fire evacuation environment to train reinforcement learning agents for evacuation planning. The environment is modelled as a graph capturing the building structure. It consists of realistic features like fire spread, uncertainty and bottlenecks. We have implemented the environment in the OpenAI gym format, to facilitate future research. We also propose a new reinforcement learning approach that entails pretraining the network weights of a DQN based agents to incorporate information on the shortest path to the exit. We achieved this by using tabular Q-learning to learn the shortest path on the building model's graph. This information is transferred to the network by deliberately overfitting it on the Q-matrix. Then, the pretrained DQN model is trained on the fire evacuation environment to generate the optimal evacuation path under time varying conditions. We perform comparisons of the proposed approach with state-of-the-art reinforcement learning algorithms like PPO, VPG, SARSA, A2C and ACKTR. The results show that our method is able to outperform state-of-the-art models by a huge margin including the original DQN based models. Finally, we test our model on a large and complex real building consisting of 91 rooms, with the possibility to move to any other room, hence giving 8281 actions. We use an attention based mechanism to deal with large action spaces. Our model achieves near optimal performance on the real world emergency environment.

• Sharma, Jivitesh; Giri, Charul; Granmo, Ole-Christoffer & Goodwin, Morten (2019). Multi-layer intrusion detection system with ExtraTrees feature selection, extreme learning machine ensemble, and softmax aggregation. EURASIP Journal on Information Security. 2019(1). doi: 10.1186/s13635-019-0098-y. Fulltekst i vitenarkiv Vis sammendrag

  Recent advances in intrusion detection systems based on machine learning have indeed outperformed other techniques, but struggle with detecting multiple classes of attacks with high accuracy. We propose a method that works in three stages. First, the ExtraTrees classifier is used to select relevant features for each type of attack individually for each (ELM). Then, an ensemble of ELMs is used to detect each type of attack separately. Finally, the results of all ELMs are combined using a softmax layer to refine the results and increase the accuracy further. The intuition behind our system is that multi-class classification is quite difficult compared to binary classification. So, we divide the multi-class problem into multiple binary classifications. We test our method on the UNSW and KDDcup99 datasets. The results clearly show that our proposed method is able to outperform all the other methods, with a high margin. Our system is able to achieve 98.24% and 99.76% accuracy for multi-class classification on the UNSW and KDDcup99 datasets, respectively. Additionally, we use the weighted extreme learning machine to alleviate the problem of imbalance in classification of attacks, which further boosts performance. Lastly, we implement the ensemble of ELMs in parallel using GPUs to perform intrusion detection in real time.

• Sharma, Jivitesh; Matheussen, Bernt Viggo; Glimsdal, Sondre & Granmo, Ole-Christoffer (2019). Hydropower Optimization Using Split-Window, Meta-Heuristic and Genetic Algorithms, 2019 18th IEEE International Conference On Machine Learning And Applications (ICMLA). IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781728145501. doi: 10.1109/ICMLA.2019.00153. Fulltekst i vitenarkiv
• Sharma, Jivitesh; Granmo, Ole-Christoffer & Goodwin, Morten (2018). Deep CNN-ELM Hybrid Models for Fire Detection in Images. Lecture Notes in Computer Science (LNCS). ISSN 0302-9743. LNCS 11141, s. 245–259. doi: 10.1007/978-3-030-01424-7_25. Fulltekst i vitenarkiv
• Sharma, Jivitesh; Granmo, Ole-Christoffer; Goodwin, Morten & Fidje, Jahn Thomas (2017). Deep convolutional neural networks for fire detection in images. Communications in Computer and Information Science (CCIS). ISSN 1865-0929. 744, s. 183–193. doi: 10.1007/978-3-319-65172-9_16. Fulltekst i vitenarkiv

Se alle arbeider i NVA

• Przybysz, Emilia; Bhattarai, Bimal; Persia, Cosimo Damiano; Ozaki, Ana; Granmo, Ole-Christoffer & Sharma, Jivitesh (2023). Verifying Properties of Tsetlin Machines. Fulltekst i vitenarkiv
• Borgersen, Karl Audun Kagnes; Goodwin, Morten & Sharma, Jivitesh (2023). Automating Tagging for Fashion Items: A User Study. Fulltekst i vitenarkiv
• Sharma, Jivitesh (2018). Deep neural network system for fire emergency: Detection, analysis and evacuation. Advances in Robotics & Automation. doi: 10.4172/2168-9695-C4-023. Fulltekst i vitenarkiv
• Sharma, Jivitesh (2018). Convolutional Neural Networks: The Emerging Technology for Medical Imaging and Gene Expression Inference. Fulltekst i vitenarkiv
• Sharma, Jivitesh (2018). Deep Neural Network System for Fire Emergencies: Detection, Analysis and Evacuation. Fulltekst i vitenarkiv
• Sharma, Jivitesh (2020). Advances in Deep Learning Towards Fire Emergency Application : Novel Architectures, Techniques and Applications of Neural Networks. Universitetet i Agder. ISSN 9788284270111. Fulltekst i vitenarkiv Vis sammendrag

  Deep Learning has been successfully used in various applications, and recently, there has been an increasing interest in applying deep learning in emergency management. However, there are still many significant challenges that limit the use of deep learning in the latter application domain. In this thesis, we address some of these challenges and propose novel deep learning methods and architectures. The challenges we address fall in these three areas of emergency management: Detection of the emergency (fire), Analysis of the situation without human intervention and finally Evacuation Planning. In this thesis, we have used computer vision tasks of image classification and semantic segmentation, as well as sound recognition, for detection and analysis. For evacuation planning, we have used deep reinforcement learning.

Se alle arbeider i NVA