Deep neural networks often fail to generalize outside of their training distribution, in particular when only a single data domain is available during training. While test-time adaptation has yielded encouraging results in this setting, we argue that, to reach further improvements, these approaches should be combined with training procedure modifications aiming to learn a more diverse set of patterns. Indeed, test-time adaptation methods usually have to rely on a limited representation because of the shortcut learning phenomenon: only a subset of the available predictive patterns is learned with standard training. In this paper, we first show that the combined use of existing training-time strategies, and test-time batch normalization, a simple adaptation method, does not always improve upon the test-time adaptation alone on the PACS benchmark. Furthermore, experiments on Office-Home show that very few training-time methods improve upon standard training, with or without test-time batch normalization. We therefore propose a novel approach using a pair of classifiers and a shortcut patterns avoidance loss that mitigates the shortcut learning behavior by reducing the generalization ability of the secondary classifier, using the additional shortcut patterns avoidance loss that encourages the learning of samples specific patterns. The primary classifier is trained normally, resulting in the learning of both the natural and the more complex, less generalizable, features. Our experiments show that our method improves upon the state-of-the-art results on both benchmarks and benefits the most to test-time batch normalization.
Research in underwater communication is rapidly becoming attractive due to its various modern applications. An efficient mechanism to secure such communication is via physical layer security. In this paper, we propose a novel physical layer authentication (PLA) mechanism in underwater acoustic communication networks where we exploit the position/location of the transmitter nodes to achieve authentication. We perform transmitter position estimation from the received signals at reference nodes deployed at fixed positions in a predefined underwater region. We use time of arrival (ToA) estimation and derive the distribution of inherent uncertainty in the estimation. Next, we perform binary hypothesis testing on the estimated position to decide whether the transmitter node is legitimate or malicious. We then provide closed-form expressions of false alarm rate and missed detection rate resulted from binary hypothesis testing. We validate our proposal via simulation results, which demonstrate errors' behavior against the link quality, malicious node location, and receiver operating characteristic (ROC) curves. We also compare our results with the performance of previously proposed fingerprint mechanisms for PLA in underwater acoustic communication networks, for which we show a clear advantage of using the position as a fingerprint in PLA.
1. Deep neural networks (DNN) have become a central class of algorithms for regression and classification tasks. Although some packages exist that allow users to specify DNN in R, those are rather limited in their functionality. Most current deep learning applications therefore rely on one of the major deep learning frameworks, PyTorch or TensorFlow, to build and train DNN. However, using these frameworks requires substantially more training and time than comparable regression or machine learning packages in the R environment. 2. Here, we present cito, an user-friendly R package for deep learning. cito allows R users to specify deep neural networks in the familiar formula syntax used by most modeling functions in R. In the background, cito uses torch to fit the models, taking advantage of all the numerical optimizations of the torch library, including the ability to switch between training models on CPUs or GPUs. Moreover, cito includes many user-friendly functions for predictions and an explainable Artificial Intelligence (xAI) pipeline for the fitted models. 3. We showcase a typical analysis pipeline using cito, including its built-in xAI features to explore the trained DNN, by building a species distribution model of the African elephant. 4. In conclusion, cito provides a user-friendly R framework to specify, deploy and interpret deep neural networks based on torch. The current stable CRAN version mainly supports fully connected DNNs, but it is planned that future versions will also include CNNs and RNNs.
The global pandemic situation has severely affected all countries. As a result, almost all countries had to adjust to online technologies to continue their processes. In addition, Sri Lanka is yearly spending ten billion on elections. We have examined a proper way of minimizing the cost of hosting these events online. To solve the existing problems and increase the time potency and cost reduction we have used IoT and ML-based technologies. IoT-based data will identify, register, and be used to secure from fraud, while ML algorithms manipulate the election data and produce winning predictions, weather-based voters attendance, and election violence. All the data will be saved in cloud computing and a standard database to store and access the data. This study mainly focuses on four aspects of an E-voting system. The most frequent problems across the world in E-voting are the security, accuracy, and reliability of the systems. E-government systems must be secured against various cyber-attacks and ensure that only authorized users can access valuable, and sometimes sensitive information. Being able to access a system without passwords but using biometric details has been there for a while now, however, our proposed system has a different approach to taking the credentials, processing, and combining the images, reformatting and producing the output, and tracking. In addition, we ensure to enhance e-voting safety. While ML-based algorithms use different data sets and provide predictions in advance.
We give the first efficient algorithm for learning halfspaces in the testable learning model recently defined by Rubinfeld and Vasilyan (2023). In this model, a learner certifies that the accuracy of its output hypothesis is near optimal whenever the training set passes an associated test, and training sets drawn from some target distribution -- e.g., the Gaussian -- must pass the test. This model is more challenging than distribution-specific agnostic or Massart noise models where the learner is allowed to fail arbitrarily if the distributional assumption does not hold. We consider the setting where the target distribution is Gaussian (or more generally any strongly log-concave distribution) in $d$ dimensions and the noise model is either Massart or adversarial (agnostic). For Massart noise our tester-learner runs in polynomial time and outputs a hypothesis with error $\mathsf{opt} + \epsilon$, which is information-theoretically optimal. For adversarial noise our tester-learner has error $\tilde{O}(\mathsf{opt}) + \epsilon$ and runs in quasipolynomial time. Prior work on testable learning ignores the labels in the training set and checks that the empirical moments of the covariates are close to the moments of the base distribution. Here we develop new tests of independent interest that make critical use of the labels and combine them with the moment-matching approach of Gollakota et al. (2023). This enables us to simulate a variant of the algorithm of Diakonikolas et al. (2020) for learning noisy halfspaces using nonconvex SGD but in the testable learning setting.
Hyperspectral imaging has the potential to improve intraoperative decision making if tissue characterisation is performed in real-time and with high-resolution. Hyperspectral snapshot mosaic sensors offer a promising approach due to their fast acquisition speed and compact size. However, a demosaicking algorithm is required to fully recover the spatial and spectral information of the snapshot images. Most state-of-the-art demosaicking algorithms require ground-truth training data with paired snapshot and high-resolution hyperspectral images, but such imagery pairs with the exact same scene are physically impossible to acquire in intraoperative settings. In this work, we present a fully unsupervised hyperspectral image demosaicking algorithm which only requires exemplar snapshot images for training purposes. We regard hyperspectral demosaicking as an ill-posed linear inverse problem which we solve using a deep neural network. We take advantage of the spectral correlation occurring in natural scenes to design a novel inter spectral band regularisation term based on spatial gradient consistency. By combining our proposed term with standard regularisation techniques and exploiting a standard data fidelity term, we obtain an unsupervised loss function for training deep neural networks, which allows us to achieve real-time hyperspectral image demosaicking. Quantitative results on hyperspetral image datasets show that our unsupervised demosaicking approach can achieve similar performance to its supervised counter-part, and significantly outperform linear demosaicking. A qualitative user study on real snapshot hyperspectral surgical images confirms the results from the quantitative analysis. Our results suggest that the proposed unsupervised algorithm can achieve promising hyperspectral demosaicking in real-time thus advancing the suitability of the modality for intraoperative use.
3D transesophageal echocardiography (3DTEE), is the recommended method for diagnosing mitral regurgitation (MR). 3DTEE provides a high-quality 3D image of the mitral valve (MV), allowing for precise segmentation and measurement of the regurgitant valve anatomy. However, manual TEE segmentations are time-consuming and prone to intra-operator variability, affecting the reliability of the measurements. To address this, we developed a fully automated pipeline using a 3D convolutional neural network (CNN) to segment MV substructures (annulus, anterior leaflet, and posterior leaflet) and quantify MV anatomy. The 3D CNN, based on a multi-decoder residual U-Net architecture, was trained and tested on a dataset comprising 100 3DTEE images with corresponding segmentations. Within the pipeline, a custom algorithm refines the CNN-based segmentations and extracts MV models, from which anatomical landmarks and features are quantified. The accuracy of the proposed method was assessed using Dice score and mean surface distance (MSD) against ground truth segmentations, and the extracted anatomical parameters were compared against a semiautomated commercial software TomTec Image Arena. The trained 3D CNN achieved an average Dice score of 0.79 and MSD of 0.47 mm for the combined segmentation of the annulus, anterior and posterior leaflet. The proposed CNN architecture outperformed a baseline residual U-Net architecture in MV substructure segmentation, and the refinement of the predicted annulus segmentation improved MSD by 8.36%. The annular and leaflet linear measurements differed by less than 7.94 mm and 3.67 mm, respectively, compared to the 3D measurements obtained with TomTec Image Arena. The proposed pipeline was faster than the commercial software, with a modeling time of 12.54 s and a quantification time of 54.42 s.
Many model-based reinforcement learning (RL) algorithms can be viewed as having two phases that are iteratively implemented: a learning phase where the model is approximately learned and a planning phase where the learned model is used to derive a policy. In the case of standard MDPs, the learning problem can be solved using either value iteration or policy iteration. However, in the case of zero-sum Markov games, there is no efficient policy iteration algorithm; e.g., it has been shown in Hansen et al. (2013) that one has to solve Omega(1/(1-alpha)) MDPs, where alpha is the discount factor, to implement the only known convergent version of policy iteration. Another algorithm for Markov zero-sum games, called naive policy iteration, is easy to implement but is only provably convergent under very restrictive assumptions. Prior attempts to fix naive policy iteration algorithm have several limitations. Here, we show that a simple variant of naive policy iteration for games converges, and converges exponentially fast. The only addition we propose to naive policy iteration is the use of lookahead in the policy improvement phase. This is appealing because lookahead is anyway often used in RL for games. We further show that lookahead can be implemented efficiently in linear Markov games, which are the counterpart of the linear MDPs and have been the subject of much attention recently. We then consider multi-agent reinforcement learning which uses our algorithm in the planning phases, and provide sample and time complexity bounds for such an algorithm.
Interleaving is an online evaluation approach for information retrieval systems that compares the effectiveness of ranking functions in interpreting the users' implicit feedback. Previous work such as Hofmann et al (2011) has evaluated the most promising interleaved methods at the time, on uniform distributions of queries. In the real world, ordinarily, there is an unbalanced distribution of repeated queries that follows a long-tailed users' search demand curve. The more a query is executed, by different users (or in different sessions), the higher the probability of collecting implicit feedback (interactions/clicks) on the related search results. This paper first aims to replicate the Team Draft Interleaving accuracy evaluation on uniform query distributions and then focuses on assessing how this method generalizes to long-tailed real-world scenarios. The reproducibility work raised interesting considerations on how the winning ranking function for each query should impact the overall winner for the entire evaluation. Based on what was observed, we propose that not all the queries should contribute to the final decision in equal proportion. As a result of these insights, we designed two variations of the $\Delta_{AB}$ score winner estimator that assign to each query a credit based on statistical hypothesis testing. To replicate, reproduce and extend the original work, we have developed from scratch a system that simulates a search engine and users' interactions from datasets from the industry. Our experiments confirm our intuition and show that our methods are promising in terms of accuracy, sensitivity, and robustness to noise.
Background: Synthetic computed tomography (sCT) has been proposed and increasingly clinically adopted to enable magnetic resonance imaging (MRI)-based radiotherapy. Deep learning (DL) has recently demonstrated the ability to generate accurate sCT from fixed MRI acquisitions. However, MRI protocols may change over time or differ between centres resulting in low-quality sCT due to poor model generalisation. Purpose: investigating domain randomisation (DR) to increase the generalisation of a DL model for brain sCT generation. Methods: CT and corresponding T1-weighted MRI with/without contrast, T2-weighted, and FLAIR MRI from 95 patients undergoing RT were collected, considering FLAIR the unseen sequence where to investigate generalisation. A ``Baseline'' generative adversarial network was trained with/without the FLAIR sequence to test how a model performs without DR. Image similarity and accuracy of sCT-based dose plans were assessed against CT to select the best-performing DR approach against the Baseline. Results: The Baseline model had the poorest performance on FLAIR, with mean absolute error (MAE)=106$\pm$20.7 HU (mean$\pm\sigma$). Performance on FLAIR significantly improved for the DR model with MAE=99.0$\pm$14.9 HU, but still inferior to the performance of the Baseline+FLAIR model (MAE=72.6$\pm$10.1 HU). Similarly, an improvement in $\gamma$-pass rate was obtained for DR vs Baseline. Conclusions: DR improved image similarity and dose accuracy on the unseen sequence compared to training only on acquired MRI. DR makes the model more robust, reducing the need for re-training when applying a model on sequences unseen and unavailable for retraining.