PELICAN is a novel permutation equivariant and Lorentz invariant or covariant aggregator network designed to overcome common limitations found in architectures applied to particle physics problems. Compared to many approaches that use non-specialized architectures that neglect underlying physics principles and require very large numbers of parameters, PELICAN employs a fundamentally symmetry group-based architecture that demonstrates benefits in terms of reduced complexity, increased interpretability, and raw performance. We present a comprehensive study of the PELICAN algorithm architecture in the context of both tagging (classification) and reconstructing (regression) Lorentz-boosted top quarks, including the difficult task of specifically identifying and measuring the $W$-boson inside the dense environment of the Lorentz-boosted top-quark hadronic final state. We also extend the application of PELICAN to the tasks of identifying quark-initiated vs.~gluon-initiated jets, and a multi-class identification across five separate target categories of jets. When tested on the standard task of Lorentz-boosted top-quark tagging, PELICAN outperforms existing competitors with much lower model complexity and high sample efficiency. On the less common and more complex task of 4-momentum regression, PELICAN also outperforms hand-crafted, non-machine learning algorithms. We discuss the implications of symmetry-restricted architectures for the wider field of machine learning for physics.

The potential surge of causal reasoning in AI models toward AGI is imminent, given the impending saturation of observation-based applications in fields like image and language processing. It is both critical and underrecognized that the essence of causality lies in the temporal nonlinearity (i.e., dynamics) of causal effects. Capturing such featured causal representations is key to realizing AGI. This paper advocates for a thorough reevaluation and potential overhaul of existing causal inference theories and the traditional learning paradigm, which predominantly relies on the observational i.i.d assumption. Our aim is to align these theories and methodologies with the intrinsic demands of AGI development. We introduce a novel Relation-Oriented paradigm for relationship modeling, and the Relation-Indexed Representation Learning (RIRL) method as its foundational implementation. Extensive experiments confirm RIRL's efficacy in autonomously capturing dynamical effects.

Research on adversarial robustness is primarily focused on image and text data. Yet, many scenarios in which lack of robustness can result in serious risks, such as fraud detection, medical diagnosis, or recommender systems often do not rely on images or text but instead on tabular data. Adversarial robustness in tabular data poses two serious challenges. First, tabular datasets often contain categorical features, and therefore cannot be tackled directly with existing optimization procedures. Second, in the tabular domain, algorithms that are not based on deep networks are widely used and offer great performance, but algorithms to enhance robustness are tailored to neural networks (e.g. adversarial training). In this paper, we tackle both challenges. We present a method that allows us to train adversarially robust deep networks for tabular data and to transfer this robustness to other classifiers via universal robust embeddings tailored to categorical data. These embeddings, created using a bilevel alternating minimization framework, can be transferred to boosted trees or random forests making them robust without the need for adversarial training while preserving their high accuracy on tabular data. We show that our methods outperform existing techniques within a practical threat model suitable for tabular data.

Gradient-based sampling algorithms have demonstrated their effectiveness in text generation, especially in the context of controlled text generation. However, there exists a lack of theoretically grounded and principled approaches for this task. In this paper, we take an important step toward building a principled approach for sampling from language models with gradient-based methods. We use discrete distributions given by language models to define densities and develop an algorithm based on Hamiltonian Monte Carlo to sample from them. We name our gradient-based technique Structured Voronoi Sampling (SVS). In an experimental setup where the reference distribution is known, we show that the empirical distribution of SVS samples is closer to the reference distribution compared to alternative sampling schemes. Furthermore, in a controlled generation task, SVS is able to generate fluent and diverse samples while following the control targets significantly better than other methods.

Recently, end-to-end neural diarization (EEND) is introduced and achieves promising results in speaker-overlapped scenarios. In EEND, speaker diarization is formulated as a multi-label prediction problem, where speaker activities are estimated independently and their dependency are not well considered. To overcome these disadvantages, we employ the power set encoding to reformulate speaker diarization as a single-label classification problem and propose the overlap-aware EEND (EEND-OLA) model, in which speaker overlaps and dependency can be modeled explicitly. Inspired by the success of two-stage hybrid systems, we further propose a novel Two-stage OverLap-aware Diarization framework (TOLD) by involving a speaker overlap-aware post-processing (SOAP) model to iteratively refine the diarization results of EEND-OLA. Experimental results show that, compared with the original EEND, the proposed EEND-OLA achieves a 14.39% relative improvement in terms of diarization error rates (DER), and utilizing SOAP provides another 19.33% relative improvement. As a result, our method TOLD achieves a DER of 10.14% on the CALLHOME dataset, which is a new state-of-the-art result on this benchmark to the best of our knowledge.

Estimating optimal dynamic policies from offline data is a fundamental problem in dynamic decision making. In the context of causal inference, the problem is known as estimating the optimal dynamic treatment regime. Even though there exists a plethora of methods for estimation, constructing confidence intervals for the value of the optimal regime and structural parameters associated with it is inherently harder, as it involves non-linear and non-differentiable functionals of unknown quantities that need to be estimated. Prior work resorted to sub-sample approaches that can deteriorate the quality of the estimate. We show that a simple soft-max approximation to the optimal treatment regime, for an appropriately fast growing temperature parameter, can achieve valid inference on the truly optimal regime. We illustrate our result for a two-period optimal dynamic regime, though our approach should directly extend to the finite horizon case. Our work combines techniques from semi-parametric inference and $g$-estimation, together with an appropriate triangular array central limit theorem, as well as a novel analysis of the asymptotic influence and asymptotic bias of softmax approximations.

Reinforcement Learning (RL) has emerged as a highly effective technique for addressing various scientific and applied problems. Despite its success, certain complex tasks remain challenging to be addressed solely with a single model and algorithm. In response, ensemble reinforcement learning (ERL), a promising approach that combines the benefits of both RL and ensemble learning (EL), has gained widespread popularity. ERL leverages multiple models or training algorithms to comprehensively explore the problem space and possesses strong generalization capabilities. In this study, we present a comprehensive survey on ERL to provide readers with an overview of recent advances and challenges in the field. Firstly, we provide an introduction to the background and motivation for ERL. Secondly, we conduct a detailed analysis of strategies such as model selection and combination that have been successfully implemented in ERL. Subsequently, we explore the application of ERL, summarize the datasets, and analyze the algorithms employed. Finally, we outline several open questions and discuss future research directions of ERL. By offering guidance for future scientific research and engineering applications, this survey significantly contributes to the advancement of ERL.

Unsupervised domain adaptation uses source data from different distributions to solve the problem of classifying data from unlabeled target domains. However, conventional methods require access to source data, which often raise concerns about data privacy. In this paper, we consider a more practical but challenging setting where the source domain data is unavailable and the target domain data is unlabeled. Specifically, we address the domain discrepancy problem from the perspective of contrastive learning. The key idea of our work is to learn a domain-invariant feature by 1) performing clustering directly in the original feature space with nearest neighbors; 2) constructing truly hard negative pairs by extended neighbors without introducing additional computational complexity; and 3) combining noise-contrastive estimation theory to gain computational advantage. We conduct careful ablation studies and extensive experiments on three common benchmarks: VisDA, Office-Home, and Office-31. The results demonstrate the superiority of our methods compared with other state-of-the-art works.

Link prediction on graphs is a fundamental problem. Subgraph representation learning approaches (SGRLs), by transforming link prediction to graph classification on the subgraphs around the links, have achieved state-of-the-art performance in link prediction. However, SGRLs are computationally expensive, and not scalable to large-scale graphs due to expensive subgraph-level operations. To unlock the scalability of SGRLs, we propose a new class of SGRLs, that we call Scalable Simplified SGRL (S3GRL). Aimed at faster training and inference, S3GRL simplifies the message passing and aggregation operations in each link's subgraph. S3GRL, as a scalability framework, accommodates various subgraph sampling strategies and diffusion operators to emulate computationally-expensive SGRLs. We propose multiple instances of S3GRL and empirically study them on small to large-scale graphs. Our extensive experiments demonstrate that the proposed S3GRL models scale up SGRLs without significant performance compromise (even with considerable gains in some cases), while offering substantially lower computational footprints (e.g., multi-fold inference and training speedup).

This paper presents the "Uncertainty-aware Mixture of Experts" (uMoE), a novel approach designed to address aleatoric uncertainty in the training of predictive models based on Neural Networks (NNs). While existing methods primarily focus on managing uncertainty during infer-ence, uMoE integrates uncertainty directly into the train-ing process. The uMoE approach adopts a "Divide and Conquer" paradigm to partition the uncertain input space into more manageable subspaces. It consists of Expert components, each trained solely on the portion of input uncertainty corresponding to their subspace. On top of the Experts, a Gating Unit, guided by additional infor-mation about the distribution of uncertain inputs across these subspaces, learns to weight the Experts to minimize deviations from the ground truth. Our results highlight that uMoE significantly outperforms baseline methods in handling data uncertainty. Furthermore, we conducted a robustness analysis, illustrating its capability to adapt to varying levels of uncertainty and suggesting optimal threshold parameters. This innovative approach holds wide applicability across diverse data-driven domains, in-cluding biomedical signal processing, autonomous driv-ing, and production quality control.