Department Publications

Test-time prompt tuning (TPT) has emerged as a promising technique for adapting large vision-language models (VLMs) to unseen tasks without relying on labeled data. However, the lack of dispersion between textual features can hurt calibration performance, which raises concerns about VLMs' reliability, trustworthiness, and safety. Current TPT approaches primarily focus on improving prompt calibration by either maximizing average textual feature dispersion or enforcing orthogonality constraints to encourage angular separation. However, these methods may not always have optimal angular separation between class-wise textual features, which implies overlooking the critical role of angular diversity. To address this, we propose A-TPT, a novel TPT framework that introduces angular diversity to encourage uniformity in the distribution of normalized textual features induced by corresponding learnable prompts …

The emergence of 5G technology has transformed network architecture, with network slicing technology standing out as a crucial feature for managing diverse and demanding applications. This innovative technology allows the creation of multiple virtual networks within a single physical infrastructure, each dedicated to specific needs such as latency, bandwidth, and reliability. These capabilities are vital for supporting services like enhanced mobile broadband, ultra-reliable low-latency communications, and massive machine-type communications. Predicting the optimal configuration of network slices is challenging due to the complex and dynamic nature of 5G environments. While traditional methods often lack the necessary accuracy, machine learning models have shown promise in addressing this challenge. However, the adoption of these models introduces the challenge of interpretability, as understanding how …

Material that constitute aggregate depend on their packing efficiency that is defined by the shape distribution of aggregates. Geometrical shape descriptors would not capture morphological aspects of different dimensional scales and frequencies, failing to wholistically numerically represent shape in material prediction models. Fourier shape descriptors are used to analyse single particle morphology, but efficiency to characterize shape of lump samples has not been assessed. This study analyses aggregates milled for different number of revolutions (0, 200, 500 and 1000) in a Los Angeles Abrasion Value instrument, that contrives morphological alterations of different scales. Means of zonal frequency components amplitudes (form, angularity and texture) are statistically different for all classes with more than 99 % confidence. Support Vector Machine and K Nearest Neighbour algorithms classified pairwise …

This manuscript introduces NeuroDetect, a model-free deep learning-based signal detection framework tailored for phase-modulated wireless systems with low-resolution analog-to-digital converters (ADCs). The proposed framework eliminates the need for explicit channel state information, which is typically difficult to acquire under coarse quantization. NeuroDetect utilizes a neural network architecture to learn the nonlinear relationship between quantized received signals and transmitted symbols directly from data. It achieves near-optimum performance, within a worst-case 12% margin of the maximum likelihood detector that assumes perfect channel knowledge. We rigorously investigate the interplay between ADC resolution and detection accuracy, introducing novel penalty metrics that quantify the effects of both quantization and learning errors. Our results shed light on the design trade-offs between ADC resolution and detection accuracy, providing future directions for developing energy-efficient high-speed and wideband wireless systems.

Image-to-point cross-modal learning has emerged to address the scarcity of large-scale 3D datasets in 3D representation learning. However, current methods that leverage 2D data often result in large, slow-to-train models, making them computationally expensive and difficult to deploy in resource-constrained environments. The architecture design of such models is therefore critical, determining their performance, memory footprint, and compute efficiency. The Joint-embedding Predictive Architecture (JEPA) has gained wide popularity in self-supervised learning for its simplicity and efficiency, but has been under-explored in cross-modal settings, partly due to the misconception that masking is intrinsic to JEPA. In this light, we propose CrossJEPA, a simple Cross-modal Joint Embedding Predictive Architecture that harnesses the knowledge of an image foundation model and trains a predictor to infer embeddings of specific rendered 2D views from corresponding 3D point clouds, thereby introducing a JEPA-style pretraining strategy beyond masking. By conditioning the predictor on cross-domain projection information, CrossJEPA purifies the supervision signal from semantics exclusive to the target domain. We further exploit the frozen teacher design with a one-time target embedding caching mechanism, yielding amortized efficiency. CrossJEPA achieves a new state-of-the-art in linear probing on the synthetic ModelNet40 (94.2%) and the real-world ScanObjectNN (88.3%) benchmarks, using only 14.1M pretraining parameters (8.5M in the point encoder), and about 6 pretraining hours on a standard single GPU. These results position CrossJEPA …

Cardiovascular diseases rank among the leading causes of mortality worldwide and the early identification of diseases is of paramount importance. This work focuses on developing a novel machine learning-based framework for early detection and classification of heart murmurs by analysing phonocardiogram signals. Our heart murmur detection and classification pipeline encompasses three classification settings. We first develop a set of methods based on transfer learning to determine the existence of heart murmurs and categorize them as present, absent, or unknown. If a murmur is present it will be classified as normal or abnormal based on its clinical outcome by using 1D convolution and audio spectrogram transformers. Finally, we use Wav2Vec encoder with raw audio data and AdaBoost abstain classifier for heart murmur quality identification. Heart murmurs are categorized based on their specific attributes …

Drone swarms pose a threat to national security. There is growing interest in an “Iron Dome for America” to protect against weaponized drones, missiles and other threats. The sensing, classification and characterization of a swarm of drones-typically with hundreds of devices flying in close proximity to each other-is a challenging radar and AI/ML problem. In this paper, we report on early work on using state-of-the-art AI/ML algorithms with advanced digital signal processing towards demonstrated capabilities for UAS swarm detection and classification at the laboratory scale. An automotive grade 77 GHz FMCW radar evaluation board, and commercial (toy) micro drones are used for experimentation. The FMCW pulses use 3 GHz of bandwidth. All experiments are conducted within a short range whereas long range experiments reserved for future work.

Machine-learning-assisted cancer subtyping is a promising avenue in digital pathology. Cancer subtyping models however require careful training using expert annotations, so that they can be inferred with a degree of known certainty (or uncertainty). To this end, we introduce the concept of uncertainty awareness into a self-supervised contrastive learning model. This is achieved by computing an evidence vector at every epoch, which assesses the model's confidence in its predictions. The derived uncertainty score is then utilized as a metric to selectively label the most crucial images that require further annotation, thus iteratively refining the training process. With just 1-10% of strategically selected annotations, we attain state-of-the-art performance in cancer subtyping on benchmark datasets. Our method not only strategically guides the annotation process to minimize the need for extensive labeled datasets, but …

Water resource management in Sri Lanka is hindered by inefficient water distribution systems in reservoirs and tanks, resulting in water scarcity, wastage, and negative impacts on agriculture, hydropower generation, and drinking water. Despite the global progress in Machine Learning (ML)-based predictive models for hydrological forecasting, Sri Lanka lacks the application of such techniques in its reservoir water management systems. This work proposes a novel approach that integrates ML and Internet of Things (IoT) technologies to develop a demand prediction and smart water distribution system tailored to the Sri Lankan context. The proposed system employs a hybrid model combining Artificial Neural Networks (ANN) and Long Short-Term Memory (LSTM) networks to accurately predict reservoir inflow, release, and water levels. Sensory data from ESP32 microcontrollers is collected and transmitted to a cloud …

Neonates are highly susceptible to seizures, often leading to short or long-term neurological impairments. However, clinical manifestations of neonatal seizures are subtle and often lead to misdiagnoses. This increases the risk of prolonged, untreated seizure activity and subsequent brain injury. Continuous video electroencephalogram (cEEG) monitoring is the gold standard for seizure detection. However, this is an expensive evaluation that requires expertise and time. In this study, we propose a convolutional neural network-based model for early prediction of neonatal seizures by distinguishing between interictal and preictal states of the EEG. Our model is patient-independent, enabling generalization across multiple subjects, and utilizes mel-frequency cepstral coefficient matrices extracted from multichannel EEG and electrocardiogram (ECG) signals as input features. Trained and validated on the Helsinki neonatal EEG dataset with 10-fold cross-validation, the proposed model achieved an average accuracy of 97.52%, sensitivity of 98.31%, specificity of 96.39%, and F1-score of 97.95%, enabling accurate seizure prediction up to 30 minutes before onset. The inclusion of ECG alongside EEG improved the F1-score by 1.42%, while the incorporation of an attention mechanism yielded an additional 0.5% improvement. To enhance transparency, we incorporated SHapley Additive exPlanations (SHAP) as an explainable artificial intelligence method to interpret the model and provided localization of seizure focus using scalp plots. The overall results demonstrate the model's potential for minimally supervised deployment in neonatal intensive care …

Objective Young children and infants, especially newborns, are highly susceptible to seizures, which, if undetected and untreated, can lead to severe long-term neurological consequences. Early detection typically requires continuous electroencephalography (cEEG) monitoring in hospital settings, involving costly equipment and highly trained specialists. This study presents a low-cost, active dry-contact electrode–based, adjustable electroencephalography (EEG) headset, combined with an explainable deep learning model for seizure detection from reduced-montage EEG, and a multimodal artifact removal algorithm to enhance signal quality. Methods EEG signals were acquired via active electrodes and processed through a custom-designed analog front end for filtering and digitization. The adjustable headset was fabricated using three-dimensional printing and laser cutting to accommodate varying head sizes …

Objective Neonates are highly susceptible to seizures, which can have severe long-term consequences if undetected and left untreated. Early detection is crucial and typically requires continuous electroencephalography (EEG) monitoring in a hospital setting, which is costly, inconvenient, and requires specialized experts for diagnosis. In this work, we propose a new low-cost active dry-contact electrode-based adjustable EEG headset, a new explainable deep learning model to detect neonatal seizures, and an advanced signal processing algorithm to remove artifacts to address the key aspects that lead to the underdiagnosis of neonatal seizures. Methods EEG signals are acquired through active electrodes and processed using a custom-designed analog front end (AFE) that filters and digitizes the captured EEG signals. The adjustable headset is designed using three-dimensional (3D) printing and laser cutting to …

Semi-supervised domain generalization (SSDG) leverages a small fraction of labeled data alongside unlabeled data to enhance model generalization. Most of the existing SSDG methods rely on pseudo-labeling (PL) for unlabeled data, often assuming access to domain labels-a privilege not always available. However, domain shifts introduce domain noise, leading to inconsistent PLs that degrade model performance. Methods derived from FixMatch suffer particularly from lower PL accuracy, reducing the effectiveness of unlabeled data. To address this, we tackle the more challenging domain-label agnostic SSDG, where domain labels for unlabeled data are not available during training. First, we propose a feature modulation strategy that enhances class-discriminative features while suppressing domain-specific information. This modulation shifts features toward Similar Average Representations-a modified version …

Low-resolution analog-to-digital converters (ADCs) have emerged as a promising solution to reduce the power consumption of massive multiple-input-multiple-output (MIMO) receivers. However, allocating a fixed number of ADC bits uniformly across all receive antennas often results in suboptimal energy efficiency (EE), particularly when channel conditions vary significantly across antennas. This paper proposes a learning-based ADC bit allocation strategy that dynamically adjusts the resolution of each ADC based on the corresponding channel gain. The proposed method effectively allocates higher-resolution ADCs to antennas experiencing stronger channel gains while minimizing the resolution for weaker channels, thereby improving the overall EE. Simulation results demonstrate that the proposed method achieves near-optimal EE performance comparable to the brute-force approach but with significantly lower …