Patient trajectories from electronic health records are widely used to predict potential outcomes of treatments over time, which then allows to personalize care. Yet, existing neural methods for this purpose have a key limitation: while some adjust for time-varying confounding, these methods assume that the time series are recorded in discrete time. In other words, they are constrained to settings where measurements and treatments are conducted at fixed time steps, even though this is unrealistic in medical practice. In this work, we aim to predict potential outcomes in continuous time. The latter is of direct practical relevance because it allows for modeling patient trajectories where measurements and treatments take place at arbitrary, irregular timestamps. We thus propose a new method called stabilized continuous time inverse propensity network (SCIP-Net). For this, we further derive stabilized inverse propensity weights for robust prediction of the potential outcomes. To the best of our knowledge, our SCIP-Net is the first neural method that performs proper adjustments for time-varying confounding in continuous time.

Fairness in machine learning seeks to mitigate model bias against individuals based on sensitive features such as sex or age, often caused by an uneven representation of the population in the training data due to selection bias. Notably, bias unascribed to sensitive features is challenging to identify and typically goes undiagnosed, despite its prominence in complex high-dimensional data from fields like computer vision and molecular biomedicine. Strategies to mitigate unidentified bias and evaluate mitigation methods are crucially needed, yet remain underexplored. We introduce: (i) Diverse Class-Aware Self-Training (DCAST), model-agnostic mitigation aware of class-specific bias, which promotes sample diversity to counter confirmation bias of conventional self-training while leveraging unlabeled samples for an improved representation of the underlying population; (ii) hierarchy bias, multivariate and class-aware bias induction without prior knowledge. Models learned with DCAST showed improved robustness to hierarchy and other biases across eleven datasets, against conventional self-training and six prominent domain adaptation techniques. Advantage was largest on multi-class classification, emphasizing DCAST as a promising strategy for fairer learning in different contexts.

There is no limit to how much a robot might explore and learn, but all of that knowledge needs to be searchable and actionable. Within language research, retrieval augmented generation (RAG) has become the workhouse of large-scale non-parametric knowledge, however existing techniques do not directly transfer to the embodied domain, which is multimodal, data is highly correlated, and perception requires abstraction. To address these challenges, we introduce Embodied-RAG, a framework that enhances the foundational model of an embodied agent with a non-parametric memory system capable of autonomously constructing hierarchical knowledge for both navigation and language generation. Embodied-RAG handles a full range of spatial and semantic resolutions across diverse environments and query types, whether for a specific object or a holistic description of ambiance. At its core, Embodied-RAG's memory is structured as a semantic forest, storing language descriptions at varying levels of detail. This hierarchical organization allows the system to efficiently generate context-sensitive outputs across different robotic platforms. We demonstrate that Embodied-RAG effectively bridges RAG to the robotics domain, successfully handling over 200 explanation and navigation queries across 19 environments, highlighting its promise for general-purpose non-parametric system for embodied agents.

This paper introduces a novel spatiotemporal feature representation model designed to address the limitations of traditional methods in multidimensional time series (MTS) analysis. The proposed approach converts MTS into one-dimensional sequences of spatially evolving events, preserving the complex coupling relationships between dimensions. By employing a variable-length tuple mining method, key spatiotemporal features are extracted, enhancing the interpretability and accuracy of time series analysis. Unlike conventional models, this unsupervised method does not rely on large training datasets, making it adaptable across different domains. Experimental results from motion sequence classification validate the model's superior performance in capturing intricate patterns within the data. The proposed framework has significant potential for applications across various fields, including backend services for monitoring and optimizing IT infrastructure, medical diagnosis through continuous patient monitoring and health trend analysis, and internet businesses for tracking user behavior and forecasting sales. This work offers a new theoretical foundation and technical support for advancing time series data mining and its practical applications in human behavior recognition and other domains.

Obtaining a reliable estimate of the joint probability mass function (PMF) of a set of random variables from observed data is a significant objective in statistical signal processing and machine learning. Modelling the joint PMF as a tensor that admits a low-rank canonical polyadic decomposition (CPD) has enabled the development of efficient PMF estimation algorithms. However, these algorithms require the rank (model order) of the tensor to be specified beforehand. In real-world applications, the true rank is unknown. Therefore, an appropriate rank is usually selected from a candidate set either by observing validation errors or by computing various likelihood-based information criteria, a procedure which is computationally expensive for large datasets. This paper presents a novel Bayesian framework for estimating the joint PMF and automatically inferring its rank from observed data. We specify a Bayesian PMF estimation model and employ appropriate prior distributions for the model parameters, allowing for tuning-free rank inference via a single training run. We then derive a deterministic solution based on variational inference (VI) to approximate the posterior distributions of various model parameters. Additionally, we develop a scalable version of the VI-based approach by leveraging stochastic variational inference (SVI) to arrive at an efficient algorithm whose complexity scales sublinearly with the size of the dataset. Numerical experiments involving both synthetic data and real movie recommendation data illustrate the advantages of our VI and SVI-based methods in terms of estimation accuracy, automatic rank detection, and computational efficiency.

Gaussian graphical models (GGMs) are widely used for recovering the conditional independence structure among random variables. Recently, several key advances have been made to exploit an additional set of variables for better estimating the GGMs of the variables of interest. For example, in co-expression quantitative trait locus (eQTL) studies, both the mean expression level of genes as well as their pairwise conditional independence structure may be adjusted by genetic variants local to those genes. Existing methods to estimate covariate-adjusted GGMs either allow only the mean to depend on covariates or suffer from poor scaling assumptions due to the inherent non-convexity of simultaneously estimating the mean and precision matrix. In this paper, we propose a convex formulation that jointly estimates the covariate-adjusted mean and precision matrix by utilizing the natural parametrization of the multivariate Gaussian likelihood. This convexity yields theoretically better performance as the sparsity and dimension of the covariates grow large relative to the number of samples. We verify our theoretical results with numerical simulations and perform a reanalysis of an eQTL study of glioblastoma multiforme (GBM), an aggressive form of brain cancer.

Conventional wisdom suggests that policy gradient methods are better suited to complex action spaces than action-value methods. However, foundational studies have shown equivalences between these paradigms in small and finite action spaces (O'Donoghue et al., 2017; Schulman et al., 2017a). This raises the question of why their computational applicability and performance diverge as the complexity of the action space increases. We hypothesize that the apparent superiority of policy gradients in such settings stems not from intrinsic qualities of the paradigm, but from universal principles that can also be applied to action-value methods to serve similar functionality. We identify three such principles and provide a framework for incorporating them into action-value methods. To support our hypothesis, we instantiate this framework in what we term QMLE, for Q-learning with maximum likelihood estimation. Our results show that QMLE can be applied to complex action spaces with a controllable computational cost that is comparable to that of policy gradient methods, all without using policy gradients. Furthermore, QMLE demonstrates strong performance on the DeepMind Control Suite, even when compared to the state-of-the-art methods such as DMPO and D4PG.

We study the least-square regression problem with a two-layer fully-connected neural network, with ReLU activation function, trained by gradient flow. Our first result is a generalization result, that requires no assumptions on the underlying regression function or the noise other than that they are bounded. We operate in the neural tangent kernel regime, and our generalization result is developed via a decomposition of the excess risk into estimation and approximation errors, viewing gradient flow as an implicit regularizer. This decomposition in the context of neural networks is a novel perspective of gradient descent, and helps us avoid uniform convergence traps. In this work, we also establish that under the same setting, the trained network overfits to the data. Together, these results, establishes the first result on benign overfitting for finite-width ReLU networks for arbitrary regression functions.

Recent work developed convolutional deep kernel machines, achieving 92.7% test accuracy on CIFAR-10 using a ResNet-inspired architecture, which is SOTA for kernel methods. However, this still lags behind neural networks, which easily achieve over 94% test accuracy with similar architectures. In this work we introduce several modifications to improve the convolutional deep kernel machine's generalisation, including stochastic kernel regularisation, which adds noise to the learned Gram matrices during training. The resulting model achieves 94.5% test accuracy on CIFAR-10. This finding has important theoretical and practical implications, as it demonstrates that the ability to perform well on complex tasks like image classification is not unique to neural networks. Instead, other approaches including deep kernel methods can achieve excellent performance on such tasks, as long as they have the capacity to learn representations from data.

The Cold Posterior Effect (CPE) is a phenomenon in Bayesian Deep Learning (BDL), where tempering the posterior to a cold temperature often improves the predictive performance of the posterior predictive distribution (PPD). Although the term `CPE' suggests colder temperatures are inherently better, the BDL community increasingly recognizes that this is not always the case. Despite this, there remains no systematic method for finding the optimal temperature beyond grid search. In this work, we propose a data-driven approach to select the temperature that maximizes test log-predictive density, treating the temperature as a model parameter and estimating it directly from the data. We empirically demonstrate that our method performs comparably to grid search, at a fraction of the cost, across both regression and classification tasks. Finally, we highlight the differing perspectives on CPE between the BDL and Generalized Bayes communities: while the former primarily focuses on predictive performance of the PPD, the latter emphasizes calibrated uncertainty and robustness to model misspecification; these distinct objectives lead to different temperature preferences.