Synthetic data algorithms are widely employed in industries to generate artificial data for downstream learning tasks. While existing research primarily focuses on empirically evaluating utility of synthetic data, its theoretical understanding is largely lacking. This paper bridges the practice-theory gap by establishing relevant utility theory in a statistical learning framework. It considers two utility metrics: generalization and ranking of models trained on synthetic data. The former is defined as the generalization difference between models trained on synthetic and on real data. By deriving analytical bounds for this utility metric, we demonstrate that the synthetic feature distribution does not need to be similar as that of real data for ensuring comparable generalization of synthetic models, provided proper model specifications in downstream learning tasks. The latter utility metric studies the relative performance of models trained on synthetic data. In particular, we discover that the distribution of synthetic data is not necessarily similar as the real one to ensure consistent model comparison. Interestingly, consistent model comparison is still achievable even when synthetic responses are not well generated, as long as downstream models are separable by a generalization gap. Finally, extensive experiments on non-parametric models and deep neural networks have been conducted to validate these theoretical findings.
相關內容
ACM/IEEE第23屆模型驅動工程語言和系統國際會議，是模型驅動軟件和系統工程的首要會議系列，由ACM-SIGSOFT和IEEE-TCSE支持組織。自1998年以來，模型涵蓋了建模的各個方面，從語言和方法到工具和應用程序。模特的參加者來自不同的背景，包括研究人員、學者、工程師和工業專業人士。MODELS 2019是一個論壇，參與者可以圍繞建模和模型驅動的軟件和系統交流前沿研究成果和創新實踐經驗。今年的版本將為建模社區提供進一步推進建模基礎的機會，并在網絡物理系統、嵌入式系統、社會技術系統、云計算、大數據、機器學習、安全、開源等新興領域提出建模的創新應用以及可持續性。
官網鏈接： ·
圖片分類
·
Pyramid
·
INFORMS
·
HTTPS
·
The traditional Transformer model encounters challenges with variable-length input sequences, particularly in Hyperspectral Image Classification (HSIC), leading to efficiency and scalability concerns. To overcome this, we propose a pyramid-based hierarchical transformer (PyFormer). This innovative approach organizes input data hierarchically into segments, each representing distinct abstraction levels, thereby enhancing processing efficiency for lengthy sequences. At each level, a dedicated transformer module is applied, effectively capturing both local and global context. Spatial and spectral information flow within the hierarchy facilitates communication and abstraction propagation. Integration of outputs from different levels culminates in the final input representation. Experimental results underscore the superiority of the proposed method over traditional approaches. Additionally, the incorporation of disjoint samples augments robustness and reliability, thereby highlighting the potential of our approach in advancing HSIC. The source code is available at //github.com/mahmad00/PyFormer.
Integrating GPUs into serverless computing platforms is crucial for improving efficiency. However, existing solutions for GPU-enabled serverless computing platforms face two significant problems due to coarse-grained GPU management: long setup time and low function throughput. To address these issues, we propose SAGE, a GPU serverless framework with fast setup and high throughput. First, based on the data knowability of GPU function ahead of actual execution, SAGE first devises the parallelized function setup mechanism, which parallelizes the data preparation and context creation. In this way, SAGE achieves fast setup of GPU function invocations.Second, SAGE further proposes the sharing-based memory management mechanism, which shares the read-only memory and context memory across multiple invocations of the same function. The memory sharing mechanism avoids repeated data preparation and then unnecessary data-loading contention. As a consequence, the function throughput could be improved. Our experimental results show that SAGE reduces function duration by 11.3X and improves function density by 1.22X compared to the state-of-the-art serverless platform.
Imputation methods for dealing with incomplete data typically assume that the missingness mechanism is at random (MAR). These methods can also be applied to missing not at random (MNAR) situations, where the user specifies some adjustment parameters that describe the degree of departure from MAR. The effect of different pre-chosen values is then studied on the inferences. This paper proposes a novel imputation method, the Random Indicator (RI) method, which, in contrast to the current methodology, estimates these adjustment parameters from the data. For an incomplete variable $X$, the RI method assumes that the observed part of $X$ is normal and the probability for $X$ to be missing follows a logistic function. The idea is to estimate the adjustment parameters by generating a pseudo response indicator from this logistic function. Our method iteratively draws imputations for $X$ and the realization of the response indicator $R$, to which we refer as $\dot{R}$, for $X$. By cross-classifying $X$ by $R$ and $\dot{R}$, we obtain various properties on the distribution of the missing data. These properties form the basis for estimating the degree of departure from MAR. Our numerical simulations show that the RI method performs very well across a variety of situations. We show how the method can be used in a real life data set. The RI method is automatic and opens up new ways to tackle the problem of MNAR data.
Safety and responsibility evaluations of advanced AI models are a critical but developing field of research and practice. In the development of Google DeepMind's advanced AI models, we innovated on and applied a broad set of approaches to safety evaluation. In this report, we summarise and share elements of our evolving approach as well as lessons learned for a broad audience. Key lessons learned include: First, theoretical underpinnings and frameworks are invaluable to organise the breadth of risk domains, modalities, forms, metrics, and goals. Second, theory and practice of safety evaluation development each benefit from collaboration to clarify goals, methods and challenges, and facilitate the transfer of insights between different stakeholders and disciplines. Third, similar key methods, lessons, and institutions apply across the range of concerns in responsibility and safety - including established and emerging harms. For this reason it is important that a wide range of actors working on safety evaluation and safety research communities work together to develop, refine and implement novel evaluation approaches and best practices, rather than operating in silos. The report concludes with outlining the clear need to rapidly advance the science of evaluations, to integrate new evaluations into the development and governance of AI, to establish scientifically-grounded norms and standards, and to promote a robust evaluation ecosystem.
We present a sublinear time algorithm for computing a near optimal low-rank approximation to any positive semidefinite (PSD) Toeplitz matrix $T\in \mathbb{R}^{d\times d}$, given noisy access to its entries. In particular, given entrywise query access to $T+E$ for an arbitrary noise matrix $E\in \mathbb{R}^{d\times d}$, integer rank $k\leq d$, and error parameter $\delta>0$, our algorithm runs in time $\text{poly}(k,\log(d/\delta))$ and outputs (in factored form) a Toeplitz matrix $\widetilde{T} \in \mathbb{R}^{d \times d}$ with rank $\text{poly}(k,\log(d/\delta))$ satisfying, for some fixed constant $C$, \begin{equation*} \|T-\widetilde{T}\|_F \leq C \cdot \max\{\|E\|_F,\|T-T_k\|_F\} + \delta \cdot \|T\|_F. \end{equation*} Here $\|\cdot \|_F$ is the Frobenius norm and $T_k$ is the best (not necessarily Toeplitz) rank-$k$ approximation to $T$ in the Frobenius norm, given by projecting $T$ onto its top $k$ eigenvectors. Our result has the following applications. When $E = 0$, we obtain the first sublinear time near-relative-error low-rank approximation algorithm for PSD Toeplitz matrices, resolving the main open problem of Kapralov et al. SODA `23, whose algorithm had sublinear query complexity but exponential runtime. Our algorithm can also be applied to approximate the unknown Toeplitz covariance matrix of a multivariate Gaussian distribution, given sample access to this distribution, resolving an open question of Eldar et al. SODA `20. Our algorithm applies sparse Fourier transform techniques to recover a low-rank Toeplitz matrix using its Fourier structure. Our key technical contribution is the first polynomial time algorithm for \emph{discrete time off-grid} sparse Fourier recovery, which may be of independent interest.
We consider a Bayesian estimator of sample size (BESS) and an application to oncology dose optimization clinical trials. BESS is built upon three pillars, Sample size, Evidence from observed data, and Confidence in posterior inference. It uses a simple logic of "given the evidence from data, a specific sample size can achieve a degree of confidence in the posterior inference." The key distinction between BESS and standard sample size estimation (SSE) is that SSE, typically based on Frequentist inference, specifies the true parameters values in its calculation while BESS assumes possible outcome from the observed data. As a result, the calibration of the sample size is not based on type I or type II error rates, but on posterior probabilities. We demonstrate that BESS leads to a more interpretable statement for investigators, and can easily accommodates prior information as well as sample size re-estimation. We explore its performance in comparison to the standard SSE and demonstrate its usage through a case study of oncology optimization trial. BESS can be applied to general hypothesis tests. An R tool is available at //ccte.uchicago.edu/BESS.
Learning on big data brings success for artificial intelligence (AI), but the annotation and training costs are expensive. In future, learning on small data is one of the ultimate purposes of AI, which requires machines to recognize objectives and scenarios relying on small data as humans. A series of machine learning models is going on this way such as active learning, few-shot learning, deep clustering. However, there are few theoretical guarantees for their generalization performance. Moreover, most of their settings are passive, that is, the label distribution is explicitly controlled by one specified sampling scenario. This survey follows the agnostic active sampling under a PAC (Probably Approximately Correct) framework to analyze the generalization error and label complexity of learning on small data using a supervised and unsupervised fashion. With these theoretical analyses, we categorize the small data learning models from two geometric perspectives: the Euclidean and non-Euclidean (hyperbolic) mean representation, where their optimization solutions are also presented and discussed. Later, some potential learning scenarios that may benefit from small data learning are then summarized, and their potential learning scenarios are also analyzed. Finally, some challenging applications such as computer vision, natural language processing that may benefit from learning on small data are also surveyed.
Residual networks (ResNets) have displayed impressive results in pattern recognition and, recently, have garnered considerable theoretical interest due to a perceived link with neural ordinary differential equations (neural ODEs). This link relies on the convergence of network weights to a smooth function as the number of layers increases. We investigate the properties of weights trained by stochastic gradient descent and their scaling with network depth through detailed numerical experiments. We observe the existence of scaling regimes markedly different from those assumed in neural ODE literature. Depending on certain features of the network architecture, such as the smoothness of the activation function, one may obtain an alternative ODE limit, a stochastic differential equation or neither of these. These findings cast doubts on the validity of the neural ODE model as an adequate asymptotic description of deep ResNets and point to an alternative class of differential equations as a better description of the deep network limit.
We describe the new field of mathematical analysis of deep learning. This field emerged around a list of research questions that were not answered within the classical framework of learning theory. These questions concern: the outstanding generalization power of overparametrized neural networks, the role of depth in deep architectures, the apparent absence of the curse of dimensionality, the surprisingly successful optimization performance despite the non-convexity of the problem, understanding what features are learned, why deep architectures perform exceptionally well in physical problems, and which fine aspects of an architecture affect the behavior of a learning task in which way. We present an overview of modern approaches that yield partial answers to these questions. For selected approaches, we describe the main ideas in more detail.
Co-evolving time series appears in a multitude of applications such as environmental monitoring, financial analysis, and smart transportation. This paper aims to address the following challenges, including (C1) how to incorporate explicit relationship networks of the time series; (C2) how to model the implicit relationship of the temporal dynamics. We propose a novel model called Network of Tensor Time Series, which is comprised of two modules, including Tensor Graph Convolutional Network (TGCN) and Tensor Recurrent Neural Network (TRNN). TGCN tackles the first challenge by generalizing Graph Convolutional Network (GCN) for flat graphs to tensor graphs, which captures the synergy between multiple graphs associated with the tensors. TRNN leverages tensor decomposition to model the implicit relationships among co-evolving time series. The experimental results on five real-world datasets demonstrate the efficacy of the proposed method.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191