Efficient Training on Very Large Corpora via Gramian Estimation
We study the problem of learning similarity functions over very large corpora using neural network embedding models. These models are typically trained using SGD with sampling of random observed and unobserved pairs, with a number of samples that grows quadratically with the corpus size, making it expensive to scale to very large corpora. We propose new efficient methods to train these models without having to sample unobserved pairs. Inspired by matrix factorization, our approach relies on adding a global quadratic penalty to all pairs of examples and expressing this term as the matrix-inner-product of two generalized Gramians. We show that the gradient of this term can be efficiently computed by maintaining estimates of the Gramians, and develop variance reduction schemes to improve the quality of the estimates. We conduct large-scale experiments that show a significant improvement in training time and generalization quality compared to traditional sampling methods.
Code (0)
등록된 구현이 없습니다.
Tasks
Network EmbeddingSimilar Papers 제목 키워드 기반
Sample Identifying Complexity of Encrypted Control Systems Under Least Squares Identification
A sample identifying complexity has been introduced in the previous study to capture an adversary's estimation error of system identification. The complexity plays a crucial role in defining the security of encrypted con…
Performance-Robustness Tradeoffs in Adversarially Robust Control and Estimation
While $\mathcal{H}_\infty$ methods can introduce robustness against worst-case perturbations, their nominal performance under conventional stochastic disturbances is often drastically reduced. Though this fundamental tra…
State EstimationReal Polynomial Gram Matrices Without Real Spectral Factors
It is well known that a non-negative definite polynomial matrix (a polynomial Gramian) $G(t)$ can be written as a product of its polynomial spectral factors, $G(t) = X(t)^H X(t)$. In this paper, we give a new algebraic c…
Resolving issues of scaling for gramian based input-output pairing methods
A key problem in process control is to decide which inputs should control which outputs. There are multiple ways to solve this problem, among them using gramian based measures, which include the Hankel interaction index …
Deep Neural Networks on EEG signals to predict Attention Score using Gramian Angular Difference Field
Auditory attention is a selective type of hearing in which people focus their attention intentionally on a specific source of a sound or spoken words whilst ignoring or inhibiting other auditory stimuli. In some sense, t…
Deep LearningEEGElectroencephalogram (EEG)Time Series Analysis