Machine Learning (cs.LG)

Abstract: In this paper, we investigate Unsupervised Episode Generation methods to solve Few-Shot Node-Classification (FSNC) problem via Meta-learning without labels. Dominant meta-learning methodologies for FSNC were developed under the existence of abundant labeled nodes for training, which however may not be possible to obtain in the real-world. Although few studies have been proposed to tackle the label-scarcity problem, they still rely on a limited amount of labeled data, which hinders the full utilization of the information of all nodes in a graph. Despite the effectiveness of Self-Supervised Learning (SSL) approaches on FSNC without labels, they mainly learn generic node embeddings without consideration on the downstream task to be solved, which may limit its performance. In this work, we propose unsupervised episode generation methods to benefit from their generalization ability for FSNC tasks while resolving label-scarcity problem. We first propose a method that utilizes graph augmentation to generate training episodes called g-UMTRA, which however has several drawbacks, i.e., 1) increased training time due to the computation of augmented features and 2) low applicability to existing baselines. Hence, we propose Neighbors as Queries (NaQ), which generates episodes from structural neighbors found by graph diffusion. Our proposed methods are model-agnostic, that is, they can be plugged into any existing graph meta-learning models, while not sacrificing much of their performance or sometimes even improving them. We provide theoretical insights to support why our unsupervised episode generation methodologies work, and extensive experimental results demonstrate the potential of our unsupervised episode generation methods for graph meta-learning towards FSNC problems.

Abstract: This paper is a contribution to the reproducibility challenge in the field of machine learning, specifically addressing the issue of certifying the robustness of neural networks (NNs) against adversarial perturbations. The proposed Double Sampling Randomized Smoothing (DSRS) framework overcomes the limitations of existing methods by using an additional smoothing distribution to improve the robustness certification. The paper provides a clear manifestation of DSRS for a generalized family of Gaussian smoothing and a computationally efficient method for implementation. The experiments on MNIST and CIFAR-10 demonstrate the effectiveness of DSRS, consistently certifying larger robust radii compared to other methods. Also various ablations studies are conducted to further analyze the hyperparameters and effect of adversarial training methods on the certified radius by the proposed framework.

Abstract: Deep neural networks (DNNs) have made significant progress, but often suffer from fairness issues, as deep models typically show distinct accuracy differences among certain subgroups (e.g., males and females). Existing research addresses this critical issue by employing fairness-aware loss functions to constrain the last-layer outputs and directly regularize DNNs. Although the fairness of DNNs is improved, it is unclear how the trained network makes a fair prediction, which limits future fairness improvements. In this paper, we investigate fairness from the perspective of decision rationale and define the parameter parity score to characterize the fair decision process of networks by analyzing neuron influence in various subgroups. Extensive empirical studies show that the unfair issue could arise from the unaligned decision rationales of subgroups. Existing fairness regularization terms fail to achieve decision rationale alignment because they only constrain last-layer outputs while ignoring intermediate neuron alignment. To address the issue, we formulate the fairness as a new task, i.e., decision rationale alignment that requires DNNs' neurons to have consistent responses on subgroups at both intermediate processes and the final prediction. To make this idea practical during optimization, we relax the naive objective function and propose gradient-guided parity alignment, which encourages gradient-weighted consistency of neurons across subgroups. Extensive experiments on a variety of datasets show that our method can significantly enhance fairness while sustaining a high level of accuracy and outperforming other approaches by a wide margin.

Abstract: Node outlier detection in attributed graphs is a challenging problem for which there is no method that would work well across different datasets. Motivated by the state-of-the-art results of score-based models in graph generative modeling, we propose to incorporate them into the aforementioned problem. Our method achieves competitive results on small-scale graphs. We provide an empirical analysis of the Dirichlet energy, and show that generative models might struggle to accurately reconstruct it.

Abstract: The rapid progress of Artificial Intelligence research came with the development of increasingly complex deep learning models, leading to growing challenges in terms of computational complexity, energy efficiency and interpretability. In this study, we apply advanced network-based information filtering techniques to design a novel deep neural network unit characterized by a sparse higher-order graphical architecture built over the homological structure of underlying data. We demonstrate its effectiveness in two application domains which are traditionally challenging for deep learning: tabular data and time series regression problems. Results demonstrate the advantages of this novel design which can tie or overcome the results of state-of-the-art machine learning and deep learning models using only a fraction of parameters.

Abstract: Federated Learning (FL) has been widely concerned for it enables decentralized learning while ensuring data privacy. However, most existing methods unrealistically assume that the classes encountered by local clients are fixed over time. After learning new classes, this assumption will make the model's catastrophic forgetting of old classes significantly severe. Moreover, due to the limitation of communication cost, it is challenging to use large-scale models in FL, which will affect the prediction accuracy. To address these challenges, we propose a novel framework, Federated Enhanced Transformer (FedET), which simultaneously achieves high accuracy and low communication cost. Specifically, FedET uses Enhancer, a tiny module, to absorb and communicate new knowledge, and applies pre-trained Transformers combined with different Enhancers to ensure high precision on various tasks. To address local forgetting caused by new classes of new tasks and global forgetting brought by non-i.i.d (non-independent and identically distributed) class imbalance across different local clients, we proposed an Enhancer distillation method to modify the imbalance between old and new knowledge and repair the non-i.i.d. problem. Experimental results demonstrate that FedET's average accuracy on representative benchmark datasets is 14.1% higher than the state-of-the-art method, while FedET saves 90% of the communication cost compared to the previous method.

Abstract: In this paper, we explore the application of mean field theory, a technique from statistical physics, to deep metric learning and address the high training complexity commonly associated with conventional metric learning loss functions. By adapting mean field theory for deep metric learning, we develop an approach to design classification-based loss functions from pair-based ones, which can be considered complementary to the proxy-based approach. Applying the mean field theory to two pair-based loss functions, we derive two new loss functions, MeanFieldContrastive and MeanFieldClassWiseMultiSimilarity losses, with reduced training complexity. We extensively evaluate these derived loss functions on three image-retrieval datasets and demonstrate that our loss functions outperform baseline methods in two out of the three datasets.

Abstract: In this paper, we delve into the critical aspect of dataset quality assessment in machine learning classification tasks. Leveraging a variety of nine distinct datasets, each crafted for classification tasks with varying complexity levels, we illustrate the profound impact of dataset quality on model training and performance. We further introduce two additional datasets designed to represent specific data conditions - one maximizing entropy and the other demonstrating high redundancy. Our findings underscore the importance of appropriate feature selection, adequate data volume, and data quality in achieving high-performing machine learning models. To aid researchers and practitioners, we propose a comprehensive framework for dataset quality assessment, which can help evaluate if the dataset at hand is sufficient and of the required quality for specific tasks. This research offers valuable insights into data assessment practices, contributing to the development of more accurate and robust machine learning models.

Abstract: We examine how transformers cope with two challenges: learning basic integer arithmetic, and generalizing to longer sequences than seen during training. We find that relative position embeddings enable length generalization for simple tasks, such as addition: models trained on $5$-digit numbers can perform $15$-digit sums. However, this method fails for multiplication, and we propose train set priming: adding a few ($10$ to $50$) long sequences to the training set. We show that priming allows models trained on $5$-digit $\times$ $3$-digit multiplications to generalize to $35\times 3$ examples. We also show that models can be primed for different generalization lengths, and that the priming sample size scales as the logarithm of the training set size. Finally, we discuss potential applications of priming beyond arithmetic.

Abstract: Neural networks (NNs) are increasingly applied in safety-critical systems such as autonomous vehicles. However, they are fragile and are often ill-behaved. Consequently, their behaviors should undergo rigorous guarantees before deployment in practice. In this paper, we propose a set-boundary reachability method to investigate the safety verification problem of NNs from a topological perspective. Given an NN with an input set and a safe set, the safety verification problem is to determine whether all outputs of the NN resulting from the input set fall within the safe set. In our method, the homeomorphism property and the open map property of NNs are mainly exploited, which establish rigorous guarantees between the boundaries of the input set and the boundaries of the output set. The exploitation of these two properties facilitates reachability computations via extracting subsets of the input set rather than the entire input set, thus controlling the wrapping effect in reachability analysis and facilitating the reduction of computation burdens for safety verification. The homeomorphism property exists in some widely used NNs such as invertible residual networks (i-ResNets) and Neural ordinary differential equations (Neural ODEs), and the open map is a less strict property and easier to satisfy compared with the homeomorphism property. For NNs establishing either of these properties, our set-boundary reachability method only needs to perform reachability analysis on the boundary of the input set. Moreover, for NNs that do not feature these properties with respect to the input set, we explore subsets of the input set for establishing the local homeomorphism property and then abandon these subsets for reachability computations. Finally, some examples demonstrate the performance of the proposed method.

Abstract: Despite its success in the image domain, adversarial training does not (yet) stand out as an effective defense for Graph Neural Networks (GNNs) against graph structure perturbations. In the pursuit of fixing adversarial training (1) we show and overcome fundamental theoretical as well as practical limitations of the adopted graph learning setting in prior work; (2) we reveal that more flexible GNNs based on learnable graph diffusion are able to adjust to adversarial perturbations, while the learned message passing scheme is naturally interpretable; (3) we introduce the first attack for structure perturbations that, while targeting multiple nodes at once, is capable of handling global (graph-level) as well as local (node-level) constraints. Including these contributions, we demonstrate that adversarial training is a state-of-the-art defense against adversarial structure perturbations.

Abstract: Context modeling and recognition are crucial for adaptive mobile and ubiquitous computing. Context-awareness in mobile environments relies on prompt reactions to context changes. However, current solutions focus on limited context information processed on centralized architectures, risking privacy leakage and lacking personalization. On-device context modeling and recognition are emerging research trends, addressing these concerns. Social interactions and visited locations play significant roles in characterizing daily life scenarios. This paper proposes an unsupervised and lightweight approach to model the user's social context and locations directly on the mobile device. Leveraging the ego-network model, the system extracts high-level, semantic-rich context features from smartphone-embedded sensor data. For the social context, the approach utilizes data on physical and cyber social interactions among users and their devices. Regarding location, it prioritizes modeling the familiarity degree of specific locations over raw location data, such as GPS coordinates and proximity devices. The effectiveness of the proposed approach is demonstrated through three sets of experiments, employing five real-world datasets. These experiments evaluate the structure of social and location ego networks, provide a semantic evaluation of the proposed models, and assess mobile computing performance. Finally, the relevance of the extracted features is showcased by the improved performance of three machine learning models in recognizing daily-life situations. Compared to using only features related to physical context, the proposed approach achieves a 3% improvement in AUROC, 9% in Precision, and 5% in Recall.

Abstract: Bayesian and causal inference are fundamental processes for intelligence. Bayesian inference models observations: what can be inferred about y if we observe a related variable x? Causal inference models interventions: if we directly change x, how will y change? Predictive coding is a neuroscience-inspired method for performing Bayesian inference on continuous state variables using local information only. In this work, we go beyond Bayesian inference, and show how a simple change in the inference process of predictive coding enables interventional and counterfactual inference in scenarios where the causal graph is known. We then extend our results, and show how predictive coding can be generalized to cases where this graph is unknown, and has to be inferred from data, hence performing causal discovery. What results is a novel and straightforward technique that allows us to perform end-to-end causal inference on predictive-coding-based structural causal models, and demonstrate its utility for potential applications in machine learning.

Abstract: In recent years, data-driven reinforcement learning (RL), also known as offline RL, have gained significant attention. However, the role of data sampling techniques in offline RL has been overlooked despite its potential to enhance online RL performance. Recent research suggests applying sampling techniques directly to state-transitions does not consistently improve performance in offline RL. Therefore, in this study, we propose a memory technique, (Prioritized) Trajectory Replay (TR/PTR), which extends the sampling perspective to trajectories for more comprehensive information extraction from limited data. TR enhances learning efficiency by backward sampling of trajectories that optimizes the use of subsequent state information. Building on TR, we build the weighted critic target to avoid sampling unseen actions in offline training, and Prioritized Trajectory Replay (PTR) that enables more efficient trajectory sampling, prioritized by various trajectory priority metrics. We demonstrate the benefits of integrating TR and PTR with existing offline RL algorithms on D4RL. In summary, our research emphasizes the significance of trajectory-based data sampling techniques in enhancing the efficiency and performance of offline RL algorithms.

Abstract: The intersection of the Foundation Model (FM) and Federated Learning (FL) provides mutual benefits, presents a unique opportunity to unlock new possibilities in AI research, and address critical challenges in AI and real-world applications. FL expands the availability of data for FMs and enables computation sharing, distributing the training process and reducing the burden on FL participants. It promotes collaborative FM development, democratizing the process and fostering inclusivity and innovation. On the other hand, FM, with its enormous size, pre-trained knowledge, and exceptional performance, serves as a robust starting point for FL, facilitating faster convergence and better performance under non-iid data. Additionally, leveraging FM to generate synthetic data enriches data diversity, reduces overfitting, and preserves privacy. By examining the interplay between FL and FM, this paper aims to deepen the understanding of their synergistic relationship, highlighting the motivations, challenges, and future directions. Through an exploration of the challenges faced by FL and FM individually and their interconnections, we aim to inspire future research directions that can further enhance both fields, driving advancements and propelling the development of privacy-preserving and scalable AI systems.

Abstract: Scientific Machine Learning (SciML) has advanced recently across many different areas in computational science and engineering. The objective is to integrate data and physics seamlessly without the need of employing elaborate and computationally taxing data assimilation schemes. However, preprocessing, problem formulation, code generation, postprocessing and analysis are still time consuming and may prevent SciML from wide applicability in industrial applications and in digital twin frameworks. Here, we integrate the various stages of SciML under the umbrella of ChatGPT, to formulate CrunchGPT, which plays the role of a conductor orchestrating the entire workflow of SciML based on simple prompts by the user. Specifically, we present two examples that demonstrate the potential use of CrunchGPT in optimizing airfoils in aerodynamics, and in obtaining flow fields in various geometries in interactive mode, with emphasis on the validation stage. To demonstrate the flow of the CrunchGPT, and create an infrastructure that can facilitate a broader vision, we built a webapp based guided user interface, that includes options for a comprehensive summary report. The overall objective is to extend CrunchGPT to handle diverse problems in computational mechanics, design, optimization and controls, and general scientific computing tasks involved in SciML, hence using it as a research assistant tool but also as an educational tool. While here the examples focus in fluid mechanics, future versions will target solid mechanics and materials science, geophysics, systems biology and bioinformatics.

Abstract: We propose a novel data-driven framework for algorithmic recourse that offers users interventions to change their predicted outcome. Existing approaches to compute recourse find a set of points that satisfy some desiderata -- e.g. an intervention in the underlying causal graph, or minimizing a cost function. Satisfying these criteria, however, requires extensive knowledge of the underlying model structure, often an unrealistic amount of information in several domains. We propose a data-driven, computationally efficient approach to computing algorithmic recourse. We do so by suggesting directions in the data manifold that users can take to change their predicted outcome. We present Stepwise Explainable Paths (StEP), an axiomatically justified framework to compute direction-based algorithmic recourse. We offer a thorough empirical and theoretical investigation of StEP. StEP offers provable privacy and robustness guarantees, and outperforms the state-of-the-art on several established recourse desiderata.

Abstract: As the frontier of machine learning applications moves further into human interaction, multiple concerns arise regarding automated decision-making. Two of the most critical issues are fairness and data privacy. On the one hand, one must guarantee that automated decisions are not biased against certain groups, especially those unprotected or marginalized. On the other hand, one must ensure that the use of personal information fully abides by privacy regulations and that user identities are kept safe. The balance between privacy, fairness, and predictive performance is complex. However, despite their potential societal impact, we still demonstrate a poor understanding of the dynamics between these optimization vectors. In this paper, we study this three-way tension and how the optimization of each vector impacts others, aiming to inform the future development of safe applications. In light of claims that predictive performance and fairness can be jointly optimized, we find this is only possible at the expense of data privacy. Overall, experimental results show that one of the vectors will be penalized regardless of which of the three we optimize. Nonetheless, we find promising avenues for future work in joint optimization solutions, where smaller trade-offs are observed between the three vectors.

Abstract: Conventional Congestion Control (CC) algorithms,such as TCP Cubic, struggle in tactical environments as they misinterpret packet loss and fluctuating network performance as congestion symptoms. Recent efforts, including our own MARLIN, have explored the use of Reinforcement Learning (RL) for CC, but they often fall short of generalization, particularly in competitive, unstable, and unforeseen scenarios. To address these challenges, this paper proposes an RL framework that leverages an accurate and parallelizable emulation environment to reenact the conditions of a tactical network. We also introduce refined RL formulation and performance evaluation methods tailored for agents operating in such intricate scenarios. We evaluate our RL learning framework by training a MARLIN agent in conditions replicating a bottleneck link transition between a Satellite Communication (SATCOM) and an UHF Wide Band (UHF) radio link. Finally, we compared its performance in file transfer tasks against Transmission Control Protocol (TCP) Cubic and the default strategy implemented in the Mockets tactical communication middleware. The results demonstrate that the MARLIN RL agent outperforms both TCP and Mockets under different perspectives and highlight the effectiveness of specialized RL solutions in optimizing CC for tactical network environments.

Abstract: We consider the problem of identifying the signal shared between two one-dimensional target variables, in the presence of additional multivariate observations. Canonical Correlation Analysis (CCA)-based methods have traditionally been used to identify shared variables, however, they were designed for multivariate targets and only offer trivial solutions for univariate cases. In the context of Multi-Task Learning (MTL), various models were postulated to learn features that are sparse and shared across multiple tasks. However, these methods were typically evaluated by their predictive performance. To the best of our knowledge, no prior studies systematically evaluated models in terms of correctly recovering the shared signal. Here, we formalize the setting of univariate shared information retrieval, and propose ICM, an evaluation metric which can be used in the presence of ground-truth labels, quantifying 3 aspects of the learned shared features. We further propose Deep Canonical Information Decomposition (DCID) - a simple, yet effective approach for learning the shared variables. We benchmark the models on a range of scenarios on synthetic data with known ground-truths and observe DCID outperforming the baselines in a wide range of settings. Finally, we demonstrate a real-life application of DCID on brain Magnetic Resonance Imaging (MRI) data, where we are able to extract more accurate predictors of changes in brain regions and obesity. The code for our experiments as well as the supplementary materials are available at https://github.com/alexrakowski/dcid

Abstract: Importance sampling is a central idea underlying off-policy prediction in reinforcement learning. It provides a strategy for re-weighting samples from a distribution to obtain unbiased estimates under another distribution. However, importance sampling weights tend to exhibit extreme variance, often leading to stability issues in practice. In this work, we consider a broader class of importance weights to correct samples in off-policy learning. We propose the use of $\textit{value-aware importance weights}$ which take into account the sample space to provide lower variance, but still unbiased, estimates under a target distribution. We derive how such weights can be computed, and detail key properties of the resulting importance weights. We then extend several reinforcement learning prediction algorithms to the off-policy setting with these weights, and evaluate them empirically.

Abstract: Large language models (LLMs) have shown promise in proving formal theorems using proof assistants such as Lean. However, existing methods are difficult to reproduce or build on, due to private code, data, and large compute requirements. This has created substantial barriers to research on machine learning methods for theorem proving. This paper removes these barriers by introducing LeanDojo: an open-source Lean playground consisting of toolkits, data, models, and benchmarks. LeanDojo extracts data from Lean and enables interaction with the proof environment programmatically. It contains fine-grained annotations of premises in proofs, providing valuable data for premise selection: a key bottleneck in theorem proving. Using this data, we develop ReProver (Retrieval-Augmented Prover): the first LLM-based prover that is augmented with retrieval for selecting premises from a vast math library. It is inexpensive and needs only one GPU week of training. Our retriever leverages LeanDojo's program analysis capability to identify accessible premises and hard negative examples, which makes retrieval much more effective. Furthermore, we construct a new benchmark consisting of 96,962 theorems and proofs extracted from Lean's math library. It features challenging data split requiring the prover to generalize to theorems relying on novel premises that are never used in training. We use this benchmark for training and evaluation, and experimental results demonstrate the effectiveness of ReProver over non-retrieval baselines and GPT-4. We thus provide the first set of open-source LLM-based theorem provers without any proprietary datasets and release it under a permissive MIT license to facilitate further research.

Abstract: State-of-the-art neural algorithmic reasoners make use of message passing in graph neural networks (GNNs). But typical GNNs blur the distinction between the definition and invocation of the message function, forcing a node to send messages to its neighbours at every layer, synchronously. When applying GNNs to learn to execute dynamic programming algorithms, however, on most steps only a handful of the nodes would have meaningful updates to send. One, hence, runs the risk of inefficiencies by sending too much irrelevant data across the graph -- with many intermediate GNN steps having to learn identity functions. In this work, we explicitly separate the concepts of node state update and message function invocation. With this separation, we obtain a mathematical formulation that allows us to reason about asynchronous computation in both algorithms and neural networks.

Abstract: Despite recent advances in synthetic data generation, the scientific community still lacks a unified consensus on its usefulness. It is commonly believed that synthetic data can be used for both data exchange and boosting machine learning (ML) training. Privacy-preserving synthetic data generation can accelerate data exchange for downstream tasks, but there is not enough evidence to show how or why synthetic data can boost ML training. In this study, we benchmarked ML performance using synthetic tabular data for four use cases: data sharing, data augmentation, class balancing, and data summarization. We observed marginal improvements for the balancing use case on some datasets. However, we conclude that there is not enough evidence to claim that synthetic tabular data is useful for ML training.

Abstract: Effective resistance (ER) is an attractive way to interrogate the structure of graphs. It is an alternative to computing the eigenvectors of the graph Laplacian. One attractive application of ER is to point clouds, i.e. graphs whose vertices correspond to IID samples from a distribution over a metric space. Unfortunately, it was shown that the ER between any two points converges to a trivial quantity that holds no information about the graph's structure as the size of the sample increases to infinity. In this study, we show that this trivial solution can be circumvented by considering a region-based ER between pairs of small regions rather than pairs of points and by scaling the edge weights appropriately with respect to the underlying density in each region. By keeping the regions fixed, we show analytically that the region-based ER converges to a non-trivial limit as the number of points increases to infinity. Namely the ER on a metric space. We support our theoretical findings with numerical experiments.

Abstract: Learning about diagnostic features and related clinical information from dental radiographs is important for dental research. However, the lack of expert-annotated data and convenient search tools poses challenges. Our primary objective is to design a search tool that uses a user's query for oral-related research. The proposed framework, Contrastive LAnguage Image REtrieval Search for dental research, Dental CLAIRES, utilizes periapical radiographs and associated clinical details such as periodontal diagnosis, demographic information to retrieve the best-matched images based on the text query. We applied a contrastive representation learning method to find images described by the user's text by maximizing the similarity score of positive pairs (true pairs) and minimizing the score of negative pairs (random pairs). Our model achieved a hit@3 ratio of 96% and a Mean Reciprocal Rank (MRR) of 0.82. We also designed a graphical user interface that allows researchers to verify the model's performance with interactions.

Abstract: This paper introduces SparseOptimizer, a novel deep learning optimizer that exploits Moreau-Yosida regularization to naturally induce sparsity in large language models such as BERT, ALBERT and GPT. Key to the design of SparseOptimizer is an embedded shrinkage operator, which imparts sparsity directly within the optimization process. This operator, backed by a sound theoretical framework, includes an analytical solution, thereby reinforcing the optimizer's robustness and efficacy. Crucially, SparseOptimizer's plug-and-play functionality eradicates the need for code modifications, making it a universally adaptable tool for a wide array of large language models. Empirical evaluations on benchmark datasets such as GLUE, RACE, SQuAD1, and SQuAD2 confirm that SparseBERT and SparseALBERT, when sparsified using SparseOptimizer, achieve performance comparable to their dense counterparts, BERT and ALBERT, while significantly reducing their parameter count. Further, this work proposes an innovative optimizer-compiler co-design strategy, demonstrating the potential of inference acceleration (\textbf{3.37x}, \textbf{6.30x}, and \textbf{7.15x} in comparison with Pytorch, TensorFlow, and LLVM generic compile, respectively) in SparseBERT when paired with an appropriately designed compiler. This study represents a significant step forward in the evolution of efficient, scalable, and high-performing large language models, setting a precedent for future exploration and optimization in this domain. The SparseOptimizer code and SparseALBERT model will be made available upon paper acceptance.

Abstract: Variational Autoencoders and their many variants have displayed impressive ability to perform dimensionality reduction, often achieving state-of-the-art performance. Many current methods however, struggle to learn good representations in High Dimensional, Low Sample Size (HDLSS) tasks, which is an inherently challenging setting. We address this challenge by using an ensemble of lightweight VAEs to learn posteriors over subsets of the feature-space, which get aggregated into a joint posterior in a novel divide-and-conquer approach. Specifically, we present an alternative factorisation of the joint posterior that induces a form of implicit data augmentation that yields greater sample efficiency. Through a series of experiments on eight real-world datasets, we show that our method learns better latent representations in HDLSS settings, which leads to higher accuracy in a downstream classification task. Furthermore, we verify that our approach has a positive effect on disentanglement and achieves a lower estimated Total Correlation on learnt representations. Finally, we show that our approach is robust to partial features at inference, exhibiting little performance degradation even with most features missing.