… , scaling kernel machines to large data … large-scale general kernel models, which are a
generalization of kernel machines that decouples the model and data, allowing training on large …

… best model obtains very competitive results among the stateof-the-arts with similar model …
on ADE20K, which shows excellent scalability towards large-scale applications. We believe …

… We establish that employing a few large kernels, instead of stacking … large kernels to build
up a deeper model, but we argue that a large-kernel ConvNet may not benefit from more large …

… design biases the VAN more toward the shape of the … the large kernel in LSKA-based VAN
encode more shape information and less texture compared to ViTs and previous large kernel …

… of kernels to be combined, and helps for automatic model selection, … for string kernels, allowing
us to train a string kernel SVM on a … We integrated multiple kernel learning in our machine …

… kernel machines that perform very well over a wide range of input dimensions and sample
sizes. This thesis aims to contribute towards bridging the gap between kernel … and large scale …

… the model’s parameters converge towards optimal values over time, influenced by factors
such as learning rate, gradient behavior, and the chosen optimization algorithm). This helps us …

… ×17×17 kernel size on Waymo 3D object detection. For the first time, we show that large
kernels are feasible and essential for 3D visual tasks. Our code and models is available at github…

… a large computational overhead. To overcome this, we propose novel approaches to
employing the large kernel, … ure 4, It can be shown that large kernels lean towards lowfrequency …

… are difficult for conventional GP models. • We study Occam’s … based model selection, which
we show is biased towards under-… step towards building probabilistic kernel machines that en…