This work has been submitted to the IEEE for possible publication. Abstract--Autonomous racing provides a controlled environment for testing the software and hardware of autonomous vehicles operating at their performance limits. Competitive interactions between multiple autonomous racecars however introduce challenging and potentially dangerous scenarios. Accurate and consistent vehicle detection and tracking is crucial for overtaking maneuvers, and low-latency sensor processing is essential to respond quickly to hazardous situations. This paper presents the LiDAR-based perception algorithms deployed on Team PoliMOVE's autonomous racecar, which won multiple competitions in the Indy Autonomous Challenge series. Our Vehicle Detection and Tracking pipeline is composed of a novel fast Point Cloud Segmentation technique and a specific Vehicle Pose Estimation methodology, together with a variable-step Multi-Figure 1. Team PoliMOVE's Dallara AV21 "MinerVa" defending from an Dallara AV21 "MinerVa" which won first place in all three In this work, we build an online algorithm for reliable I. UTONOMOUS RACING allows for safe testing of an autonomous vehicle's full software and hardware stack fully observing the target's 2D pose, tracking its motion at the limits of its performance in a controlled environment. Point Cloud segmentation algorithm capable of processing in Providing this kind of testing environment is one of the main parallel the three LiDAR sensors mounted on the vehicle, a goals of the Indy Autonomous Challenge (IAC), the first multivehicle multi-hypothesis L-shape fitting technique for a racing vehicle competition series for level 4 autonomous racecars.

Feature selection is a critical step in the analysis of high-dimensional data, where the number of features often vastly exceeds the number of samples. Effective feature selection not only improves model performance and interpretability but also reduces computational costs and mitigates the risk of overfitting. In this context, we propose a novel feature selection method for high-dimensional data, based on the well-known permutation feature importance approach, but extending it to evaluate subsets of attributes rather than individual features. This extension more effectively captures how interactions among features influence model performance. The proposed method employs a multi-objective evolutionary algorithm to search for candidate feature subsets, with the objectives of maximizing the degradation in model performance when the selected features are shuffled, and minimizing the cardinality of the feature subset. The effectiveness of our method has been validated on a set of 24 publicly available high-dimensional datasets for classification and regression tasks, and compared against 9 well-established feature selection methods designed for high-dimensional problems, including the conventional permutation feature importance method. The results demonstrate the ability of our approach in balancing accuracy and computational efficiency, providing a powerful tool for feature selection in complex, high-dimensional datasets.

In this work, we evaluate annotator disagreement in Word-in-Context (WiC) tasks exploring the relationship between contextual meaning and disagreement as part of the CoMeDi shared task competition. While prior studies have modeled disagreement by analyzing annotator attributes with single-sentence inputs, this shared task incorporates WiC to bridge the gap between sentence-level semantic representation and annotator judgment variability. We describe three different methods that we developed for the shared task, including a feature enrichment approach that combines concatenation, element-wise differences, products, and cosine similarity, Euclidean and Manhattan distances to extend contextual embedding representations, a transformation by Adapter blocks to obtain task-specific representations of contextual embeddings, and classifiers of varying complexities, including ensembles. The comparison of our methods demonstrates improved performance for methods that include enriched and task-specfic features. While the performance of our method falls short in comparison to the best system in subtask 1 (OGWiC), it is competitive to the official evaluation results in subtask 2 (DisWiC).

Parkinson's disease (PD) is a progressive neurodegenerative disorder that impacts motor functions and speech characteristics This study focuses on differentiating individuals with Parkinson's disease from healthy controls through the extraction and classification of speech features. Patients were further divided into 2 groups. Med On represents the patient with medication, while Med Off represents the patient without medication. The dataset consisted of patients and healthy individuals who read a predefined text using the H1N Zoom microphone in a suitable recording environment at F{\i}rat University Neurology Department. Speech recordings from PD patients and healthy controls were analyzed, and 19 key features were extracted, including jitter, luminance, zero-crossing rate (ZCR), root mean square (RMS) energy, entropy, skewness, and kurtosis.These features were visualized in graphs and statistically evaluated to identify distinctive patterns in PD patients. Using MATLAB's Classification Learner toolbox, several machine learning classification algorithm models were applied to classify groups and significant accuracy rates were achieved. The accuracy of our 3-layer artificial neural network architecture was also compared with classical machine learning algorithms. This study highlights the potential of noninvasive voice analysis combined with machine learning for early detection and monitoring of PD patients. Future research can improve diagnostic accuracy by optimizing feature selection and exploring advanced classification techniques.

We introduce Feasible Learning (FL), a sample-centric learning paradigm where models are trained by solving a feasibility problem that bounds the loss for each training sample. In contrast to the ubiquitous Empirical Risk Minimization (ERM) framework, which optimizes for average performance, FL demands satisfactory performance on every individual data point. Since any model that meets the prescribed performance threshold is a valid FL solution, the choice of optimization algorithm and its dynamics play a crucial role in shaping the properties of the resulting solutions. In particular, we study a primal-dual approach which dynamically re-weights the importance of each sample during training. To address the challenge of setting a meaningful threshold in practice, we introduce a relaxation of FL that incorporates slack variables of minimal norm. Our empirical analysis, spanning image classification, age regression, and preference optimization in large language models, demonstrates that models trained via FL can learn from data while displaying improved tail behavior compared to ERM, with only a marginal impact on average performance.

Clarity: While I can understand the statements in this paper, I think the presentation can be much improved. For example: - As RPU learning implies PAC learning, is there really a need to present Algorithm 1 and Theorem 3.5? Aren't we already happy with Theorem 4.11? - in Algorithm 1, Threshold(S) is only informally defined, and an elaboration is needed. I think basically, the algorithm can successfully approximately recover b if it can find two neighboring and - examples? Also, what is the active learning algorithm used here?

The reviewer are unanimous in their support of accepting this paper. The paper makes an important contribution to the literature on learning with label queries and comparison queries, showing that comparison queries dramatically improve the query complexity of learning (nonhomogeneous) halfspaces under distribution assumptions, and furthermore the results even hold in the more-challenging "RPU" model (where the predictor must never be wrong, but may abstain with epsilon probability). The approach stems from general principles, and may lead to further follow-up works.

Weaknesses: My main concerns are whether this work makes an impactful contribution to the type of problems of interest to the NeurIPS community. The problem of high-dimensional integration is of course of vital importance in many areas of machine learning, appearing centrally for example in Bayesian inference/model selection, graphical models, and the training of latent variable generative models, and many of us would welcome an addition to the toolkit of dealing with such beasts. Unfortunately, this paper makes only minimal effort to motivate the relevance of the proposed QMC construction to these settings. An application to GAN/VAEs does briefly appear in the supplementary, but with quite cursory quantification of performance; showing sharper generated images is not consistent with the rigorous aims and tone of the paper. For a NeurIPS audience, I consider it essential to include a comparison against established sampling algorithms such as Sequential Monte Carlo.

Dear Authors: I read your rebuttal. I do indeed understand the point of your paper. I also agree that solving linear equations is not a good example of something kernel methods can't do. The'isotonic regression' algorithm for learning a ReLU is a simple SGD algorithm that uses a straight-through estimator. The high-level message of your paper is that there are problems that gradient-based methods can solve, but kernel methods cannot.

This paper proposes a zeroth-order adaptive momentum method for black-box optimization, by approximating the stochastic gradient using the forward difference of two function values at a random unit direction. The paper also shows the convergence analysis in terms of Mahalanobis distance for both unconstrained and constrained nonconvex optimization with the ZO-AdaMM, which results in sublinear convergence rates that are roughly a factor of the square root of dimension worse than that of the first-order ZO-AdaMM, as well as for constrained convex optimization. The proposed scheme is quite interesting, which solves the (non)convex optimization in a new perspective, and somewhat provides new insight to the adaptive momentum methods. In particular, the paper provides a formal conclusion that the Euclidean projection may results in non-convergence issue in stochastic optimization. The paper also shows the applications to black-box adversarial attacks problems and validate the method by comparing it with other ZO methods.