Training machine learning models on decentralized private data via federated learning (FL) poses two key challenges: communication efficiency and privacy protection. In this work, we address these challenges within the trusted aggregator model by introducing a novel approach called the Communication-Efficient and Privacy-Adaptable Mechanism (CEPAM), achieving both objectives simultaneously. In particular, CEPAM leverages the rejection-sampled universal quantizer (RSUQ), a construction of randomized vector quantizer whose resulting distortion is equivalent to a prescribed noise, such as Gaussian or Laplace noise, enabling joint differential privacy and compression. Moreover, we analyze the trade-offs among user privacy, global utility, and transmission rate of CEPAM by defining appropriate metrics for FL with differential privacy and compression. Our CEPAM provides the additional benefit of privacy adaptability, allowing clients and the server to customize privacy protection based on required accuracy and protection. We assess CEPAM's utility performance using MNIST dataset, demonstrating that CEPAM surpasses baseline models in terms of learning accuracy.

Whole slide images (WSIs) are high-resolution, gigapixel sized images that pose significant computational challenges for traditional machine learning models due to their size and heterogeneity.In this paper, we present a scalable and efficient methodology for WSI classification by leveraging patch-based feature extraction, clustering, and Fisher vector encoding. Initially, WSIs are divided into fixed size patches, and deep feature embeddings are extracted from each patch using a pre-trained convolutional neural network (CNN). These patch-level embeddings are subsequently clustered using K-means clustering, where each cluster aggregates semantically similar regions of the WSI. To effectively summarize each cluster, Fisher vector representations are computed by modeling the distribution of patch embeddings in each cluster as a parametric Gaussian mixture model (GMM). The Fisher vectors from each cluster are concatenated into a high-dimensional feature vector, creating a compact and informative representation of the entire WSI. This feature vector is then used by a classifier to predict the WSI's diagnostic label. Our method captures local and global tissue structures and yields robust performance for large-scale WSI classification, demonstrating superior accuracy and scalability compared to other approaches.

Stochastic iterative algorithms, including stochastic gradient descent (SGD) and stochastic gradient Langevin dynamics (SGLD), are widely utilized for optimization and sampling in large-scale and high-dimensional problems in machine learning, statistics, and engineering. Numerous works have bounded the parameter error in, and characterized the uncertainty of, these approximations. One common approach has been to use scaling limit analyses to relate the distribution of algorithm sample paths to a continuous-time stochastic process approximation, particularly in asymptotic setups. Focusing on the univariate setting, in this paper, we build on previous work to derive non-asymptotic functional approximation error bounds between the algorithm sample paths and the Ornstein-Uhlenbeck approximation using an infinite-dimensional version of Stein's method of exchangeable pairs. We show that this bound implies weak convergence under modest additional assumptions and leads to a bound on the error of the variance of the iterate averages of the algorithm. Furthermore, we use our main result to construct error bounds in terms of two common metrics: the L\'{e}vy-Prokhorov and bounded Wasserstein distances. Our results provide a foundation for developing similar error bounds for the multivariate setting and for more sophisticated stochastic approximation algorithms.

Gaussian Mixture Models (GMMs) range among the most frequently used machine learning models. However, training large, general GMMs becomes computationally prohibitive for datasets with many data points $N$ of high-dimensionality $D$. For GMMs with arbitrary covariances, we here derive a highly efficient variational approximation, which is integrated with mixtures of factor analyzers (MFAs). For GMMs with $C$ components, our proposed algorithm significantly reduces runtime complexity per iteration from $\mathcal{O}(NCD^2)$ to a complexity scaling linearly with $D$ and remaining constant w.r.t. $C$. Numerical validation of this theoretical complexity reduction then shows the following: the distance evaluations required for the entire GMM optimization process scale sublinearly with $NC$. On large-scale benchmarks, this sublinearity results in speed-ups of an order-of-magnitude compared to the state-of-the-art. As a proof of concept, we train GMMs with over 10 billion parameters on about 100 million images, and observe training times of approximately nine hours on a single state-of-the-art CPU.

Sequences provide a remarkably general way to represent and process information. This powerful abstraction has placed sequence modeling at the center of modern deep learning applications, inspiring numerous architectures from transformers to recurrent networks. While this fragmented development has yielded powerful models, it has left us without a unified framework to understand their fundamental similarities and explain their effectiveness. We present a unifying framework motivated by an empirical observation: effective sequence models must be able to perform associative recall. Our key insight is that memorizing input tokens through an associative memory is equivalent to performing regression at test-time. This regression-memory correspondence provides a framework for deriving sequence models that can perform associative recall, offering a systematic lens to understand seemingly ad-hoc architectural choices. We show numerous recent architectures -- including linear attention models, their gated variants, state-space models, online learners, and softmax attention -- emerge naturally as specific approaches to test-time regression. Each architecture corresponds to three design choices: the relative importance of each association, the regressor function class, and the optimization algorithm. This connection leads to new understanding: we provide theoretical justification for QKNorm in softmax attention, and we motivate higher-order generalizations of softmax attention. Beyond unification, our work unlocks decades of rich statistical tools that can guide future development of more powerful yet principled sequence models.

Blockchain technology, with implications in the financial domain, offers data in the form of large-scale transaction networks. Analyzing transaction networks facilitates fraud detection, market analysis, and supports government regulation. Despite many graph representation learning methods for transaction network analysis, we pinpoint two salient limitations that merit more investigation. Existing methods predominantly focus on the snapshots of transaction networks, sidelining the evolving nature of blockchain transaction networks. Existing methodologies may not sufficiently emphasize efficient, incremental learning capabilities, which are essential for addressing the scalability challenges in ever-expanding large-scale transaction networks. To address these challenges, we employed an incremental approach for random walk-based node representation learning in transaction networks. Further, we proposed a Metropolis-Hastings-based random walk mechanism for improved efficiency. The empirical evaluation conducted on blockchain transaction datasets reveals comparable performance in node classification tasks while reducing computational overhead. Potential applications include transaction network monitoring, the efficient classification of blockchain addresses for fraud detection or the identification of specialized address types within the network.

Discovering dominant patterns and exploring dynamic behaviors especially critical state transitions and tipping points in high-dimensional time-series data are challenging tasks in study of real-world complex systems, which demand interpretable data representations to facilitate comprehension of both spatial and temporal information within the original data space. Here, we proposed a general and analytical ultralow-dimensionality reduction method for dynamical systems named spatial-temporal principal component analysis (stPCA) to fully represent the dynamics of a high-dimensional time-series by only a single latent variable without distortion, which transforms high-dimensional spatial information into one-dimensional temporal information based on nonlinear delay-embedding theory. The dynamics of this single variable is analytically solved and theoretically preserves the temporal property of original high-dimensional time-series, thereby accurately and reliably identifying the tipping point before an upcoming critical transition. Its applications to real-world datasets such as individual-specific heterogeneous ICU records demonstrated the effectiveness of stPCA, which quantitatively and robustly provides the early-warning signals of the critical/tipping state on each patient.

This work provides a strong contribution in that it apparently is the first work to net optimal rates (up to log factors) for SGM, and moreover, it also handles a mini-batch analysis which includes the (full) batch method as a special case. Such rates previously had been established only for the (batch) ridge regression method. My interpretation of what all the results actually show is given in the Summary. I find the current solution of relying on cross-validation for adaptation to be a bit of an inelegant cop-out (even if there is a theoretically-supported method for using it); given that several of your corollaries provide a guarantee where \zeta and \gamma enter the picture only through T *, can you provide a self-monitoring method that decides when to stop? In particular, I find the most exciting results to be Corollaries 3.3 and 3.9, as only the stopping time depends on the (unknown) capacity parameters, and so such an online stopping mechanism might be possible.

This paper proposes an original idea and theoretically appealing solutions to solve it. Quality: Its first part (Section 1, 2) is excellent, but its latter part may be a little weak. Section 3 is a little dense with numerous details and heuristics. A pseudo-code showing the overall framework may be helpful for readers. Section 4 is a little short to validate the potential effectiveness of the proposed method. A direct comparison between CKN and CNN with same architecture may be more informative for readers.

Comments (mostly on related work): Authors: "The idea of using learning to learn or meta-learning to acquire knowledge or inductive biases has a long history [Thrun and Pratt, 1998]." But the intro to this reference is muddling the waters by confusing meta-learning (which is about learning the learning algorithm itself) and transfer learning, subsuming basically everything under "meta-learning," even standard back-propagation, because it can be applied to some data set, and then may learn new data points more quickly (so this is just standard transfer learning). To my knowledge, the first work on learning general learning algorithms written in a universal programming language was published in 1987: J. Schmidhuber. Evolutionary principles in self-referential learning, or on learning how to learn: The meta-meta-... hook. Authors: "This work was built on by [Younger et al., 2001, Hochreiter et al., 2001] wherein a higher-level network act as a gradient descent procedure, with both levels trained during learning. Alternatively Schmidhuber [1992, 1993] considers networks that are able to modify their own behavior and act as an alternative to recurrent networks in meta-learning. Note, however that these earlier works do not directly address the transfer of a learned training procedure to novel problem instances and instead focus on adaptivity in the online setting."