Medical image segmentation models often struggle to generalize across different domains due to various reasons. Domain Generalization (DG) methods overcome this either through representation learning or data augmentation (DA). While representation learning methods seek domain-invariant features, they often rely on ad-hoc techniques and lack formal guarantees. DA methods, which enrich model representations through synthetic samples, have shown comparable or superior performance to representation learning approaches. We propose LangDAug, a novel Langevin Data Augmentation for multi-source domain generalization in 2D medical image segmentation. LangDAug leverages Energy-Based Models (EBMs) trained via contrastive divergence to traverse between source domains, generating intermediate samples through Langevin dynamics. Theoretical analysis shows that LangDAug induces a regularization effect, and for GLMs, it upper-bounds the Rademacher complexity by the intrinsic dimensionality of the data manifold. Through extensive experiments on Fundus segmentation and 2D MRI prostate segmentation benchmarks, we show that LangDAug outperforms state-of-the-art domain generalization methods and effectively complements existing domain-randomization approaches. The codebase for our method is available at https://github.com/backpropagator/LangDAug.

Bayesian methods provide an elegant framework for estimating parameter posteriors and quantification of uncertainty associated with probabilistic models. However, they often suffer from slow inference times. To address this challenge, Bayesian Pseudo-Coresets (BPC) have emerged as a promising solution. BPC methods aim to create a small synthetic dataset, known as pseudo-coresets, that approximates the posterior inference achieved with the original dataset. This approximation is achieved by optimizing a divergence measure between the true posterior and the pseudo-coreset posterior. Various divergence measures have been proposed for constructing pseudo-coresets, with forward Kullback-Leibler (KL) divergence being the most successful. However, using forward KL divergence necessitates sampling from the pseudo-coreset posterior, often accomplished through approximate Gaussian variational distributions. Alternatively, one could employ Markov Chain Monte Carlo (MCMC) methods for sampling, but this becomes challenging in high-dimensional parameter spaces due to slow mixing. In this study, we introduce a novel approach for constructing pseudo-coresets by utilizing contrastive divergence. Importantly, optimizing contrastive divergence eliminates the need for approximations in the pseudo-coreset construction process. Furthermore, it enables the use of finite-step MCMC methods, alleviating the requirement for extensive mixing to reach a stationary distribution. To validate our method’s effectiveness, we conduct extensive experiments on multiple datasets, demonstrating its superiority over existing BPC techniques. Our implementation is available at https://github.com/backpropagator/BPC-CD .

Class imbalance is a common phenomenon in multiple application domains such as healthcare, where the sample occurrence of one or few class categories is more prevalent in the dataset than the rest. This work addresses the class-imbalance issue by proposing an over-sampling method for the minority classes in the latent space of a Regularized Auto-Encoder (RAE). Specifically, we construct a latent space by maximizing the conditional data likelihood using an Encoder-Decoder structure, such that oversampling through convex combinations of latent samples preserves the class identity. A jointly-trained linear classifier that separates convexly coupled latent vectors from different classes is used to impose this property on the AE’s latent space. Further, the aforesaid linear classifier is used for final classification without retraining. We theoretically show that our method can achieve a low variance risk estimate compared to naive oversampling methods and is robust to overfitting. We conduct several experiments on benchmark datasets and show that our method outperforms the existing oversampling techniques for handling class imbalance. The code of the proposed method is available at: https://github.com/arnabkmondal/oversamplingrae.

Location estimation in an indoor Internet-of-Things (IoT) environment is a challenging task due to multipath signals and obstacles that cause shadowing and fading effects, and change the received signal power considerably. Most of the existing path-loss-based localization methods assume only a lognormal shadowing model and ignore small scale fading effects. This article considers a generic combined lognormal shadowing and Rayleigh fading model for efficient localization of smart devices in an indoor IoT environment. In particular, the maximum likelihood estimate of the location and path-loss exponent (PLE), and Cramer-Rao lower bound (CRLB) are derived. The localization parameters are estimated using a novel adaptive mini-batch gradient ascent method that maximizes the log-likelihood function with an appropriate batch size based on the convergence factor. Hence, the proposed method addresses the challenge of an arbitrary selection of a fixed batch size for a gradient ascent method by utilizing this convergence factor. Performance evaluation by a simulation study and real experiments from an indoor IoT testbed provide a more accurate joint estimation of model parameters and smart device localization.

Location-based services assume significant importance in the Internet of Things (IoT) based systems. In the scenarios where the satellite signals are not available or weak, the Global Positioning System (GPS) accuracy degrades sharply. Therefore, opportunistic signals can be utilized for smart device localization. In this paper, we propose a smart device localization method using residual neural networks. The proposed network is generic and performs smart device localization using opportunistic signals such as Wireless Fidelity (Wi-Fi), geomagnetic, temperature, pressure, humidity, and light signals in the IoT network. Additionally, the proposed method addresses the two significant challenges in IoT based smart device localization, which are noise and device heterogeneity. The experiments are performed on three real datasets of different opportunistic signals. Results show that the proposed method is robust to noise, and a significant improvement in the localization accuracy is obtained as compared to the state-of-the-art localization methods.