Segment Anything Model 2 (SAM2) has demonstrated state-of-the-art performance in image/video object segmentation across many domains, but its large encoder makes it challenging for resource-constrained devices or real-time applications. One solution to this problem is to carry out knowledge distillation from the bulky encoder to a lightweight encoder, but this can result in degraded performance. In this work, we investigate multi-teacher distillation to mitigate performance degradation for distilled segmentation models. Using several foundation teacher models, our multi-teacher distilled models achieve 3.2 times speedup during end-to-end inference compared to SAM2 while achieving the best results of 74.4 and 71.1 (72.1 and 69.6 for single-teacher distillation) mIoU on the COCO and LVIS image segmentation datasets, as well as showing competitive results on video segmentation. Our results show that multi-teacher distillation offers a powerful solution for efficient image/video segmentation, while also maintaining compelling performance.

Conventional Time Series Classification (TSC) methods are often black boxes that obscure inherent interpretation of their decision-making processes. In this work, we leverage Multiple Instance Learning (MIL) to overcome this issue, and propose a new framework called MILLET: Multiple Instance Learning for Locally Explainable Time series classification. We apply MILLET to existing deep learning TSC models and show how they become inherently interpretable without compromising (and in some cases, even improving) predictive performance. We evaluate MILLET on 85 UCR TSC datasets and also present a novel synthetic dataset that is specially designed to facilitate interpretability evaluation. On these datasets, we show MILLET produces sparse explanations quickly that are of higher quality than other well-known interpretability methods. To the best of our knowledge, our work with MILLET is the first to develop general MIL methods for TSC and apply them to an extensive variety of domains.

The composition of multiple Gaussian Processes as a Deep Gaussian Process DGP enables a deep probabilistic nonparametric approach to flexibly tackle complex machine learning problems with sound quantification of uncertainty. Existing inference approaches for DGP models have limited scalability and are notoriously cumbersome to construct. In this work we introduce a novel formulation of DGPs based on random feature expansions that we train using stochastic variational inference. This yields a practical learning framework which significantly advances the state-of-the-art in inference for DGPs, and enables accurate quantification of uncertainty. We extensively showcase the scalability and performance of our proposal on several datasets with up to 8 million observations, and various DGP architectures with up to 30 hidden layers.

The computational and storage complexity of kernel machines presents the primary barrier to their scaling to large, modern, datasets. A common way to tackle the scalability issue is to use the conjugate gradient algorithm, which relieves the constraints on both storage (the kernel matrix need not be stored) and computation (both stochastic gradients and parallelization can be used). Even so, conjugate gradient is not without its own issues: the conditioning of kernel matrices is often such that conjugate gradients will have poor convergence in practice. Preconditioning is a common approach to alleviating this issue. Here we propose preconditioned conjugate gradients for kernel machines, and develop a broad range of preconditioners particularly useful for kernel matrices. We describe a scalable approach to both solving kernel machines and learning their hyperparameters. We show this approach is exact in the limit of iterations and outperforms state-of-the-art approximations for a given computational budget.