Search for: All records

BackgroundReliable and specific biomarkers that can distinguish autism spectrum disorders (ASDs) from commonly co-occurring attention-deficit/hyperactivity disorder (ADHD) are lacking, causing misses and delays in diagnosis, and reducing access to interventions and quality of life. AimsTo examine whether an innovative, brief (1-min), videogame method called Computerised Assessment of Motor Imitation (CAMI), can identify ASD-specific imitation differences compared with neurotypical children and children with ADHD. MethodThis cross-sectional study used CAMI alongside standardised parent-report (Social Responsiveness Scale, Second Edition) and observational measures of autism (Autism Diagnostic Observation Schedule-Second Edition; ADOS-2), ADHD (Conners) and motor ability (Physical and Neurological Examination for Soft Signs). The sample comprised 183 children aged 7–13 years, with ADHD (without ASD), with ASD (with and without ADHD) and who were neurotypical. ResultsRegardless of co-occurring ADHD, children with ASD showed poorer CAMI performance than neurotypical children (P< 0.0001; adjustedR²= 0.28), whereas children with ADHD and neurotypical children showed similar CAMI performance. Receiver operating curve and support vector machine analyses showed that CAMI distinguishes ASD from both neurotypical children (80% true positive rate) and children with ADHD (70% true positive rate), with a high success rate significantly above chance. Among children with ASD, poor CAMI performance was associated with increased autism traits, particularly ADOS-2 measures of social affect and restricted and repetitive behaviours (adjustedR²= 0.23), but not with ADHD traits or motor ability. ConclusionsFour levels of analyses confirm that poor imitation measured by the low-cost and scalable CAMI method specifically distinguishes ASD not only from neurotypical development, but also from commonly co-occurring ADHD. 

The advent of large pre-trained models has brought about a paradigm shift in both visual representation learning and natural language processing. However, clustering unlabeled images, as a fundamental and classic machine learning problem, still lacks an effective solution, particularly for large-scale datasets. In this paper, we propose a novel image clustering pipeline that leverages the powerful feature representation of large pre-trained models such as CLIP and cluster images effectively and efficiently at scale. We first developed a novel algorithm to estimate the number of clusters in a given dataset. We then show that the pre-trained features are significantly more structured by further optimizing the rate reduction objective. The resulting features may significantly improve the clustering accuracy, e.g., from 57\% to 66\% on ImageNet-1k. Furthermore, by leveraging CLIP's multimodality bridge between image and text, we develop a simple yet effective self-labeling algorithm that produces meaningful captions for the clusters. Through extensive experiments, we show that our pipeline works well on standard datasets such as CIFAR-10, CIFAR-100, and ImageNet-1k. It also extends to datasets that are not curated for clustering, such as LAION-Aesthetics and WikiArts. We released the code in https://github.com/LeslieTrue/CPP 

Outlier-robust estimation involves estimating some parameters (e.g., 3D rotations) from data samples in the presence of outliers, and is typically formulated as a non-convex and non-smooth problem. For this problem, the classical method called iteratively reweighted least-squares (IRLS) and its variants have shown impressive performance. This paper makes several contributions towards understanding why these algorithms work so well. First, we incorporate majorization and graduated non-convexity (GNC) into the IRLS framework and prove that the resulting IRLS variant is a convergent method for outlier-robust estimation. Moreover, in the robust regression context with a constant fraction of outliers, we prove this IRLS variant converges to the ground truth at a global linear and local quadratic rate for a random Gaussian feature matrix with high probability. Experiments corroborate our theory and show that the proposed IRLS variant converges within 5-10 iterations for typical problem instances of outlier-robust estimation, while state-of-the-art methods need at least 30 iterations. A basic implementation of our method is provided: https: //github.com/liangzu/IRLS-CVPR2023 

The problem of clustering points on a union of subspaces finds numerous applications in machine learning and computer vision, and it has been extensively studied in the past two decades. When the subspaces are low-dimensional, the problem can be formulated as a convex sparse optimization problem, for which numerous accurate, efficient and robust methods exist. When the subspaces are of high relative dimension (e.g., hyperplanes), the problem is intrinsically non-convex, and existing methods either lack theory, are computationally costly, lack robustness to outliers, or learn hyperplanes one at a time. In this paper, we propose Hyperplane ARangentment Descent (HARD), a method that robustly learns all the hyperplanes simultaneously by solving a novel non-convex non-smooth ℓ1 minimization problem. We provide geometric conditions under which the ground-truth hyperplane arrangement is a coordinate-wise minimizer of our objective. Furthermore, we devise efficient algorithms, and give conditions under which they converge to coordinate-wise minimizes. We provide empirical evidence that HARD surpasses state-of-the-art methods and further show an interesting experiment in clustering deep features on CIFAR-10. 

We advance both the theory and practice of robust ℓp-quasinorm regression for p ∈ (0,1] by using novel variants of iteratively reweighted least-squares (IRLS) to solve the underlying non-smooth problem. In the convex case, p =1, we prove that this IRLS variant converges globally at a linear rate under a mild, deterministic condition on the feature matrix called the stable range space property. In the non-convex case, p ∈ (0,1), we prove that under a similar condition, IRLS converges locally to the global minimizer at a superlinear rate of order 2−p; the rate becomes quadratic as p→0. We showcase the proposed methods in three applications: real phase retrieval, regression without correspondences, and robust face restoration. The results show that (1) IRLS can handle a larger number of outliers than other methods, (2) it is faster than competing methods at the same level of accuracy, (3) it restores a sparsely corrupted face image with satisfactory visual quality. 

Abstract Arguably, the two most popular accelerated or momentum-based optimization methods in machine learning are Nesterov’s accelerated gradient and Polyaks’s heavy ball, both corresponding to different discretizations of a particular second order differential equation with friction. Such connections with continuous-time dynamical systems have been instrumental in demystifying acceleration phenomena in optimization. Here we study structure-preserving discretizations for a certain class of dissipative (conformal) Hamiltonian systems, allowing us to analyse the symplectic structure of both Nesterov and heavy ball, besides providing several new insights into these methods. Moreover, we propose a new algorithm based on a dissipative relativistic system that normalizes the momentum and may result in more stable/faster optimization. Importantly, such a method generalizes both Nesterov and heavy ball, each being recovered as distinct limiting cases, and has potential advantages at no additional cost.