In recent years, the rapid development of deep neural networks has brought increased attention to the security and robustness of these models. While existing adversarial attack algorithms have demonstrated success in improving adversarial transferability, their performance remains suboptimal due to a lack of consideration for the discrepancies between target and source models. To address this limitation, we propose a novel method, Inverse Knowledge Distillation (IKD), designed to enhance adversarial transferability effectively. IKD introduces a distillation-inspired loss function that seamlessly integrates with gradient-based attack methods, promoting diversity in attack gradients and mitigating overfitting to specific model architectures. By diversifying gradients, IKD enables the generation of adversarial samples with superior generalization capabilities across different models, significantly enhancing their effectiveness in black-box attack scenarios.

The state-of-the-art dimensionality reduction approaches largely rely on complicated optimization procedures. On the other hand, closed-form approaches requiring merely eigen-decomposition do not have enough sophistication and nonlinearity. In this paper, we propose a novel nonlinear dimensionality reduction method -- Inverse Kernel Decomposition (IKD) -- based on an eigen-decomposition of the sample covariance matrix of data. The method is inspired by Gaussian process latent variable models (GPLVMs) and has comparable performance with GPLVMs. To deal with very noisy data with weak correlations, we propose two solutions -- blockwise and geodesic -- to make use of locally correlated data points and provide better and numerically more stable latent estimations. We use synthetic datasets and four real-world datasets to show that IKD is a better dimensionality reduction method than other eigen-decomposition-based methods, and achieves comparable performance against optimization-based methods with faster running speeds. Open-source IKD implementation in Python can be accessed at this \url{https://github.com/JerrySoybean/ikd}.

Deep neural network (DNN) models for retinopathy have estimated predictive accuracies in the mid-to-high 90%. However, the following aspects remain unaddressed: State-of-the-art models are complex and require substantial computational infrastructure to train and deploy; The reliability of predictions can vary widely. In this paper, we focus on these aspects and propose a form of iterative knowledge distillation(IKD), called IKD+ that incorporates a tradeoff between size, accuracy and reliability. We investigate the functioning of IKD+ using two widely used techniques for estimating model calibration (Platt-scaling and temperature-scaling), using the best-performing model available, which is an ensemble of EfficientNets with approximately 100M parameters. We demonstrate that IKD+ equipped with temperature-scaling results in models that show up to approximately 500-fold decreases in the number of parameters than the original ensemble without a significant loss in accuracy. In addition, calibration scores (reliability) for the IKD+ models are as good as or better than the base mode

We assume that a high-dimensional datum, like an image, is a compositional expression of a set of properties, with a complicated non-linear relationship between the datum and its properties. This paper proposes a factorial mixture prior for capturing latent properties, thereby adding structured compositionality to deep generative models. The prior treats a latent vector as belonging to Cartesian product of subspaces, each of which is quantized separately with a Gaussian mixture model. Some mixture components can be set to represent properties as observed random variables whenever labeled properties are present. Through a combination of stochastic variational inference and gradient descent, a method for learning how to infer discrete properties in an unsupervised or semi-supervised way is outlined and empirically evaluated.