2026
Vincent le Roux, Marelie H. Davel, Johan Bosman, "A deep learning framework for expert-guided sailplane performance optimisation", Intelligent Systems with Applications 31, 2026. [Paper]
Coenraad Mouton, Randle Rabe, Niklas C. Koser, Nicolai Krekiehn, Christopher Hansen, Jan-Bernd Hövener, Claus-C. Glüer, "Useful nonrobust features are ubiquitous in biomedical images", IEEE International Symposium on Biomedical Imaging (ISBI), 2026. [Paper] [ArXiv Preprint]
Ruan P. Van der Spoel, Randle Rabe, "Investigating the relationship between diversity and generalization in deep neural networks", Northern Lights Deep Learning Conference (NLDL), Proceedings of Machine Learning Research 307, 2026. [Paper]
2025
Marthinus W. Theunissen, Randle Rabe, Marelie H. Davel, "KnowIt: deep time series modeling and interpretation", arXiv preprint, 2025. [ArXiv Preprint]
Leon Freese, Marthinus W. Theunissen, "Data pruning: redundant, problematic, and interdependent samples", Southern African Conference for Artificial Intelligence Research (SACAIR), Communications in Computer and Information Science 2784, 2025. [Paper]
Simbarashe A. Ngorima, Albert S. J. Helberg, Marelie H. Davel, "Feature extraction for plant growth estimation", Southern African Conference for Artificial Intelligence Research (SACAIR), Communications in Computer and Information Science 2784, 2025. [Paper] [ArXiv Preprint]
Shelley Haupt, Bolelang Sibolla, Raymond Molapo, Lizwe Mdakane, Nicolene Fourie, "Exploring the use of data in a digital twin for the marine and coastal environment", ISPRS International Journal of Geo-Information 14(4), 2025. [Paper]
Simbarashe A. Ngorima, Albert S. J. Helberg, Marelie H. Davel, "Simplified temporal convolutional-based channel estimation for a WiFi vehicular communication channel", IEEE Wireless Africa Conference (WAC), 2025. [Paper] [ArXiv Preprint]
2024
Coenraad Mouton, Marthinus W. Theunissen, Marelie H. Davel, "Input margins can predict generalization too", AAAI Conference on Artificial Intelligence (AAAI), 2024. [Paper] [ArXiv Preprint]
Vincent le Roux, Marelie H. Davel, Johan Bosman, "Parsimonious airfoil parameterisation: a deep learning framework with bidirectional LSTM and Gaussian mixture models", Expert Systems with Applications 255, 2024. [Paper]
Coenraad Mouton, Randle Rabe, Daniël G. Haasbroek, Marthinus W. Theunissen, Hermanus L. Potgieter, Marelie H. Davel, "Is network fragmentation a useful complexity measure?", Scientific Methods for Understanding Deep Learning Workshop (SciForDL at NeurIPS), 2024. [Paper] [ArXiv Preprint]
Simbarashe A. Ngorima, Albert S. J. Helberg, Marelie H. Davel, "A data pilot-aided temporal convolutional network for channel estimation in IEEE 802.11p vehicle-to-vehicle communications", Southern Africa Telecommunication Networks and Applications Conference (SATNAC), 2024. [Paper] [ArXiv Preprint]
Simbarashe A. Ngorima, Albert S. J. Helberg, Marelie H. Davel, "Neural network-based vehicular channel estimation performance: effect of noise in the training set", Southern African Conference for Artificial Intelligence Research (SACAIR), Communications in Computer and Information Science 2326, 2024. [Paper] [ArXiv Preprint]
Simon P. Ramalepe, Thipe I. Modipa, Marelie H. Davel, "Pre-training a transformer-based generative model using a small Sepedi dataset", Southern African Conference for Artificial Intelligence Research (SACAIR), Communications in Computer and Information Science 2326, 2024. [Paper] [ArXiv Preprint]
Hermanus L. Potgieter, Coenraad Mouton, Marelie H. Davel, "Impact of batch normalization on convolutional network representations", Southern African Conference for Artificial Intelligence Research (SACAIR), Communications in Computer and Information Science 2326, 2024. [Paper] [ArXiv Preprint]
William Brooks, Marelie H. Davel, Coenraad Mouton, "Does simple trump complex? Comparing strategies for adversarial robustness in DNNs", Southern African Conference for Artificial Intelligence Research (SACAIR), Communications in Computer and Information Science 2326, 2024. [Paper]
2023
Coenraad Middel, Marelie H. Davel, "Comparing transformer-based and GBDT models on tabular data: a Rossmann Store Sales case study", Southern African Conference for Artificial Intelligence Research (SACAIR), 2023. [Paper]
Jan C. Olivier, Etienne Barnard, "Minimum phase finite impulse response filter design", IET Signal Processing 17(1), 2023. [Paper]
Marelie H. Davel, Stefan Lotz, Marthinus W. Theunissen, Almaro de Villiers, Chara Grant, Randle Rabe, Stefan Schoombie, Cleo Conacher, "Knowledge discovery in time series data", Deep Learning Indaba, 2023. [Poster]
Simbarashe A. Ngorima, Albert S. J. Helberg, Marelie H. Davel, "Sequence based deep neural networks for channel estimation in vehicular communication systems", Southern African Conference for Artificial Intelligence Research (SACAIR), Communications in Computer and Information Science 1976, 2023. [Paper]
Simon P. Ramalepe, Thipe I. Modipa, Marelie H. Davel, "The analysis of the Sepedi-English code-switched radio news corpus", Journal of the Digital Humanities Association of Southern Africa 4(1), 2023. [Paper]
Simon P. Ramalepe, Thipe I. Modipa, Marelie H. Davel, "Transformer-based text generation for code-switched Sepedi-English news", Southern African Conference for Artificial Intelligence Research (SACAIR), 2023. [Paper]
Stefan Lotz, Amore E. Nel, Robert T. Wicks, Owen W. Roberts, Nicholas E. Engelbrecht, Roelf D. Strauss, Gert J. J. Botha, Eduard P. Kontar, Alexander Pitňa, Stuart D. Bale, "The radial variation of the solar wind turbulence spectra near the kinetic break scale from Parker Solar Probe measurements", The Astrophysical Journal 942(2), 2023. [Paper]
2022
Edrich Fourie, Marelie H. Davel, Jaco Versfeld, "Neural speech processing for whale call detection", Southern African Conference for Artificial Intelligence Research (SACAIR), Communications in Computer and Information Science 1734, 2022. [Paper]
Walter Heymans, Marelie H. Davel, Charl van Heerden, "Multi-style training for South African call centre audio", Southern African Conference for Artificial Intelligence Research (SACAIR), Communications in Computer and Information Science 1551, 2022. [Paper] [ArXiv Preprint]
Walter Heymans, Marelie H. Davel, Charl van Heerden, "Efficient acoustic feature transformation in mismatched environments using a Guided-GAN", Speech Communication 143, 2022. [Paper] [ArXiv Preprint]
Thipe I. Modipa, Marelie H. Davel, "Two Sepedi-English code-switched speech corpora", Language Resources and Evaluation 56(3), 2022. [Paper]
Coenraad Mouton, Marelie H. Davel, "Exploring layerwise decision making in DNNs", Southern African Conference for Artificial Intelligence Research (SACAIR), Communications in Computer and Information Science 1551, 2022. [Paper] [ArXiv Preprint]
Andrew J. Oosthuizen, Marelie H. Davel, Albert S. J. Helberg, "Multi-layer perceptron for channel state information estimation: design considerations", Southern Africa Telecommunication Networks and Applications Conference (SATNAC), 2022. [Paper]
Andrew J. Oosthuizen, Albert S. J. Helberg, Marelie H. Davel, "Adversarial training for channel state information estimation in LTE multi-antenna systems", Southern African Conference for Artificial Intelligence Research (SACAIR), Communications in Computer and Information Science 1734, 2022. [Paper]
Marko Oosthuizen, Alwyn Hoffman, Marelie H. Davel, "A comparative study of graph neural network speed prediction during periods of congestion", International Conference on Neural Computation Theory and Applications (NCTA), 2022. [Paper]
Simon P. Ramalepe, Thipe I. Modipa, Marelie H. Davel, "The development of a Sepedi text generation model using transformers", Southern Africa Telecommunication Networks and Applications Conference (SATNAC), 2022. [Paper]
Marthinus W. Theunissen, Coenraad Mouton, Marelie H. Davel, "The missing margin: how sample corruption affects distance to the boundary measurements in ANNs", Southern African Conference for Artificial Intelligence Research (SACAIR), Communications in Computer and Information Science 1734, 2022. [Paper] [ArXiv Preprint]
Ruan van der Merwe, Gregory Newman, Etienne Barnard, "Manifold characteristics that predict downstream task performance", Workshop on Pre-training: Perspectives, Pitfalls, and Paths Forward (at ICML), 2022. [ArXiv Preprint]
2021
Daniël G. Haasbroek, Marelie H. Davel, "Exploring neural network training dynamics through binary node activations", Southern African Conference for Artificial Intelligence Research (SACAIR), 2021. [Paper]
Nuette Heyns, Etienne Barnard, "Optimising word embeddings for recognised multilingual speech", Southern African Conference for Artificial Intelligence Research (SACAIR), 2021. [Paper]
Dylan Lamprecht, Etienne Barnard, "Using a meta-model to compensate for training-evaluation mismatches", Southern African Conference for Artificial Intelligence Research (SACAIR), 2021. [Paper]
Andrew J. Oosthuizen, Marelie H. Davel, Albert S. J. Helberg, "Exploring CNN-based automatic modulation classification using small modulation sets", Southern Africa Telecommunication Networks and Applications Conference (SATNAC), 2021. [Paper]
Rhyno A. Strydom, Etienne Barnard, "Classifying recognised speech with deep neural networks", Southern African Conference for Artificial Intelligence Research (SACAIR), 2021. [Paper]
Lucas van Wyk, Marelie H. Davel, Charl van Heerden, "Unsupervised fine-tuning of speaker diarisation pipelines using silhouette coefficients", Southern African Conference for Artificial Intelligence Research (SACAIR), 2021. [Paper]
2020
Jacques P. Beukes, Stefan Lotz, Marelie H. Davel, "Pairwise networks for feature ranking of a geomagnetic storm model", South African Computer Journal 32(2), 2020. [Paper]
Marelie H. Davel, "Using summary layers to probe neural network behaviour", South African Computer Journal 32(2), 2020. [Paper]
Marelie H. Davel, Marthinus W. Theunissen, Arnold M. Pretorius, Etienne Barnard, "DNNs as layers of cooperating classifiers", AAAI Conference on Artificial Intelligence (AAAI), 2020. [Paper] [ArXiv Preprint]
Ulrike Janke, Etienne Barnard, "MuST: a case study on the development of AI systems in the South African context", Litnet Seminars and Essays, 2020. [Paper]
Coenraad Mouton, Johannes C. Myburgh, Marelie H. Davel, "Stride and translation invariance in CNNs", Southern African Conference for Artificial Intelligence Research (SACAIR), Communications in Computer and Information Science 1342, 2020. [Paper] [ArXiv Preprint]
Johannes C. Myburgh, Coenraad Mouton, Marelie H. Davel, "Tracking translation invariance in CNNs", Southern African Conference for Artificial Intelligence Research (SACAIR), Communications in Computer and Information Science 1342, 2020. [Paper] [ArXiv Preprint]
Marthinus W. Theunissen, Marelie H. Davel, Etienne Barnard, "Benign interpolation of noise in deep learning", South African Computer Journal 32(2), 2020. [Paper]
Jan W. F. Thirion, Charl van Heerden, Oluwapelumi Giwa, Marelie H. Davel, "The South African directory enquiries (SADE) name corpus", Language Resources and Evaluation 54(1), 2020. [Paper]
Arthur E. W. Venter, Marthinus W. Theunissen, Marelie H. Davel, "Pre-interpolation loss behavior in neural networks", Southern African Conference for Artificial Intelligence Research (SACAIR), Communications in Computer and Information Science 1342, 2020. [Paper] [ArXiv Preprint]
2019
Marelie H. Davel, "Activation gap generators in neural networks", South African Forum for Artificial Intelligence Research (FAIR), 2019. [Paper]
Dewald D. Krynauw, Marelie H. Davel, Stefan Lotz, "Solar flare prediction with temporal convolutional networks", South African Forum for Artificial Intelligence Research (FAIR), 2019. [Paper]
Stefan Lotz, Jacques P. Beukes, Marelie H. Davel, "Input parameter ranking for neural networks in a space weather regression problem", South African Forum for Artificial Intelligence Research (FAIR), 2019. [Paper]
Stefan Lotz, Jacques P. Beukes, Marelie H. Davel, "A neural network based method for input parameter selection", Machine Learning in Heliophysics, 2019. [Poster]
Arnold M. Pretorius, Etienne Barnard, Marelie H. Davel, "ReLU and sigmoidal activation functions", South African Forum for Artificial Intelligence Research (FAIR), 2019. [Paper]
Marthinus W. Theunissen, Marelie H. Davel, Etienne Barnard, "Insights regarding overfitting on noise in deep learning", South African Forum for Artificial Intelligence Research (FAIR), 2019. [Paper]