Latent DiTs improve data distribution recovery and inference efficiency under low-dimensional latent space.
problem Improving data distribution recovery and inference efficiency in latent DiTs.
method Investigates statistical and computational limits of latent DiTs under low-dimensional latent space assumption, deriving approximation error bounds, sample complexity, and efficient inference and training algorithms.
result Latent DiTs can bypass high dimensionality challenges and achieve almost-linear time inference and training.
LTMs use latent vectors for efficient autoregressive generation.
problem Efficient autoregressive generation in language models.
method Dual-rate optimization in variational Bayes framework.
result LTMs achieve superior sample and parameter efficiency.
Latent-IMH improves Bayesian inference for expensive operators.
problem Efficient sampling from posterior distributions in inverse problems with computationally expensive operators.
method Metropolis-Hastings independence sampler using approximate and exact operators.
result Latent-IMH outperforms existing methods in computational efficiency.
Deep NLP models benefit from underlying structures in the data---e.g., parse trees---typically extracted using off-the-shelf parsers. Recent attempts to jointly learn the latent structure encounter a tradeoff: either make factorization assumptions that limit expressiveness, or sacrifice end-to-end differentiability. Us…
New method uses Diffusion Maps for latent space modeling of dynamical systems.
problem Building reduced dynamical models from time series data.
method Two rounds of Diffusion Maps on latent coordinates, with lifting back to ambient space.
result Approximation of full state functions in reduced coordinates.
This text explores strategies for learning discrete latent structures in neural networks.
problem Learning discrete latent structures in neural networks is challenging.
method Continuous relaxation, surrogate gradients, and probabilistic estimation.
result Many latent structure learning strategies use the same fundamental building blocks but apply them differently.
This paper shows how to estimate distances in latent space of random graphs using entropic OT.
problem Estimating distances between groups of nodes in latent space of random graphs.
method Entropic Optimal Transport (OT) with stability results for perturbations of the cost matrix.
result Consistent estimation of entropic OT distances between groups of nodes in latent space.
New computational lower bounds for clustering and related problems.
problem Statistical-computational gaps in high-dimensional clustering problems.
method Investigation of low-degree polynomials in latent space models to derive lower bounds.
result New and sharper computational lower bounds for clustering, sparse clustering, and biclustering.
Computer simulations have become a popular tool of assessing complex skills such as problem-solving skills. Log files of computer-based items record the entire human-computer interactive processes for each respondent. The response processes are very diverse, noisy, and of nonstandard formats. Few generic methods have b…
Develops methods to estimate high rank tensors from noisy data.
problem Estimating high rank tensors from noisy observations.
method Generative latent variable tensor model, polynomial-time spectral algorithm.
result Achieves computationally optimal rate for signal tensor estimation.
New method identifies latent variables in cognitive models using neural networks.
problem Inference of latent variables in complex cognitive models is limited.
method Recurrent neural networks and simulation-based inference for latent variable sequences.
result Extends neural Bayes estimation to broader classes of cognitive models.
New methods improve sampling from complex dynamical models.
problem Sampling from high-dimensional, non-linear latent dynamical models is computationally challenging.
method Introduce auxiliary MCMC and Particle Gibbs samplers with improved performance and parallelisation.
result Enhanced samplers maintain performance in high-dimensional latent spaces and support parallelisation.
Autoregressive models struggle with hard-to-compute distributions, alternatives like energy-based and latent-variable models solve this.
problem Autoregressive models struggle with distributions whose next-symbol probability is hard to compute.
method Alternatives include energy-based models and latent-variable autoregressive models.
result Alternatives to autoregressive models can escape limitations of hard-to-compute distributions.
New findings suggest latent regularization is unnecessary for high-quality image generation.
problem Improving image generation quality without latent regularization.
method Investigated the effect of latent regularization on image generation using learned priors.
result In the case of a sufficiently expressive prior, latent regularization is not necessary and may harm image quality.
We introduce a Bayesian Gaussian process latent variable model that explicitly captures spatial correlations in data using a parameterized spatial kernel and leveraging structure-exploiting algebra on the model covariance matrices for computational tractability. Inference is made tractable through a collapsed variation…
A new tree-Wasserstein distance for high-dimensional data with latent feature hierarchy.
problem Finding meaningful distances between high-dimensional data samples with latent feature hierarchy.
method Proposes a new tree-Wasserstein distance (TWD) for high-dimensional data with a latent feature hierarchy, using diffusion geometry and tree decoding.
result The proposed TWD effectively recovers the latent feature hierarchy and is efficient and scalable.
A scalable GPVAE method using local adjacencies to approximate GP inference.
problem Scalability issues in exact GP inference for large-scale GPVAEs.
method Neighbour-driven approximation strategy that confines computations to nearest neighbours.
result Outperforms other GPVAE variants in predictive performance and computational efficiency.
This paper generalizes graph representation for diverse data types.
problem Representing and querying hybrid data types in a unified way.
method Introducing a directed Tensor-Typed Multi-Graph with embeddings.
result Unified representation for visual, linguistic, and auditory data.
DiCoLa recursively decomposes causal structure learning for latent variables.
problem Learning causal structures in high-dimensional settings with latent variables.
method Recursive decomposition framework for divide-and-conquer causal discovery.
result Theoretical soundness and completeness of DiCoLa framework.
Symmetric binary matrices representing relations among entities are commonly collected in many areas. Our focus is on dynamically evolving binary relational matrices, with interest being in inference on the relationship structure and prediction. We propose a nonparametric Bayesian dynamic model, which reduces dimension…
A new method for efficient inference in sequential latent-variable models.
problem Computational challenges in integrating subject-specific random effects.
method Anchored variational inference framework to approximate posterior distributions.
result The method achieves accurate estimation with significant computational gains.
SyNGLER generates synthetic networks efficiently while preserving key structural properties.
problem Efficiently generating realistic synthetic networks with preserved structural properties.
method SyNGLER uses latent space network models to learn and reconstruct node embeddings, then generates synthetic networks.
result SyNGLER produces synthetic networks that better preserve key network characteristics than existing approaches.
Ensemble decoders to capture latent space topology in deep generative models.
problem Topological mismatch between latent space geometry and data manifolds.
method Using ensembles of decoders to compute geodesics on the expected manifold.
result Ensemble approach provides a simple and reliable way to capture model uncertainty in latent space.
Estimating graphical model structure from high-dimensional and undersampled data is a fundamental problem in many scientific fields. Existing approaches, such as GLASSO, latent variable GLASSO, and latent tree models, suffer from high computational complexity and may impose unrealistic sparsity priors in some cases. We…
SKR-VAE improves VAEs for ICA with reduced computational cost.
problem Efficiently performing ICA in VAEs with large datasets.
method Structured kernel functions to avoid costly GP kernel matrix inversion.
result SKR-VAE achieves greater computational efficiency and reduced resource consumption.
Matrix factorisation methods decompose multivariate observations as linear combinations of latent feature vectors. The Indian Buffet Process (IBP) provides a way to model the number of latent features required for a good approximation in terms of regularised reconstruction error. Previous work has focussed on latent fe…
Paper introduces Latent-CLIP for efficient text-image comparison in latent space.
problem Efficiently compare text and images in latent space without costly decoding.
method Trains CLIP model in latent space, uses Latent-CLIP rewards for noise optimization, and guides generation away from harmful content.
result Latent-CLIP matches CLIP performance on text-image classification and harmful content detection.
Deep generative models are tremendously successful in learning low-dimensional latent representations that well-describe the data. These representations, however, tend to much distort relationships between points, i.e. pairwise distances tend to not reflect semantic similarities well. This renders unsupervised tasks, s…
DSVNP uses global and local latent variables for improved neural process predictions.
problem Limited expressiveness of vanilla neural processes in capturing target-specific local variation.
method Introduces DSVNP combining global and local latent variables for prediction.
result Competitive prediction performance in multi-output regression and uncertainty estimation.
A new method learns latent variable updates directly, not approximating the posterior.
problem Intractable maximum-likelihood learning for complex latent-variable models.
method Amortised learning using wake-sleep Monte-Carlo strategy.
result Demonstrated effectiveness on various complex models.
The paper establishes a nearly-sharp statistical threshold for efficient learning in Latent MDPs with separated components.
problem Learning Latent Markov Decision Processes (LMDPs) with separated components.
method The paper considers various notions of separation and establishes a nearly-sharp statistical threshold for efficient learning. It also presents a quasi-polynomial algorithm with time complexity scaling in terms of the statistical threshold under a weaker assumption of separability under the optimal policy, and a near-matching time complexity lower bound under the exponential time hypothesis.
result Establishes a nearly-sharp statistical threshold for efficient learning in Latent MDPs with separated components.
Enhances image-to-image translation using adversarial latent space.
problem Image-to-image translation task in computer vision.
method Introduces an adversarial discriminator on the latent representation to enforce similar latent space distributions.
result Significantly outperforms competing approaches on MNIST and USPS domain adaptation tasks.
SDE Matching eliminates simulation for training Latent SDEs, achieving similar performance.
problem Training Latent SDEs with adjoint sensitivity methods is computationally expensive and limited.
method SDE Matching, inspired by Score- and Flow Matching, eliminates simulation for training Latent SDEs.
result SDE Matching achieves performance comparable to adjoint sensitivity methods while reducing computational complexity.
We consider the problem of inferring a latent function in a probabilistic model of data. When dependencies of the latent function are specified by a Gaussian process and the data likelihood is complex, efficient computation often involve Markov chain Monte Carlo sampling with limited applicability to large data sets. W…
Gaussian latent tree models, or more generally, Gaussian latent forest models have Fisher-information matrices that become singular along interesting submodels, namely, models that correspond to subforests. For these singularities, we compute the real log-canonical thresholds (also known as stochastic complexities or l…
We improve autoencoder image interpolation by shaping latent space.
problem Incongruities in autoencoder interpolation leading to artifacts or unrealistic results.
method Propose a regularization technique to shape latent space to follow a smooth, locally convex manifold consistent with training images.
result Faithful interpolation between data points achieved.
Modified EAT method improves Poisson gradient estimation.
problem Challenging differentiation through Poisson-distributed latent variables.
method Exponential Arrival Time (EAT) simulation with modifications and Gumbel-SoftMax relaxation.
result Modified EAT method provides unbiased first moment and reduced second-moment bias.
Proposes iVDFM for identifying latent factors in multivariate time series.
problem Identifying latent factors in multivariate time series with structural dynamics.
method Identifiable Variational Dynamic Factor Model (iVDFM) with iVAE-style conditioning.
result Identifiable latent factors up to permutation and component-wise affine transformations.
LASER compresses recursive model activations by exploiting their low-dimensional structure.
problem Understanding and optimizing the geometric structure of recursive reasoning trajectories.
method Dynamic low-rank basis tracking via matrix-free subspace tracking with a fidelity-triggered reset mechanism.
result Recursive activations occupy a linear, low-dimensional subspace that can be compressed efficiently.
Latent force models (LFM) are principled approaches to incorporating solutions to differential equations within non-parametric inference methods. Unfortunately, the development and application of LFMs can be inhibited by their computational cost, especially when closed-form solutions for the LFM are unavailable, as is …
Sloth predicts LLM performance using latent skills across families.
problem Variations in benchmark performance due to differences in training configurations and data processing across model families.
method Sloth uses publicly available benchmark data and assumes LLM performance is driven by latent skills influenced by model size and training tokens. It exploits correlations across benchmarks to provide accurate predictions.
result Sloth predicts LLM performance accurately and offers insights into scaling behaviors for complex tasks.
Proposes LAE-EnKF for improved nonlinear data assimilation.
problem Performance of EnKF deteriorates for strongly nonlinear dynamics.
method Reformulates assimilation in a learned latent space with linear dynamics.
result LAE-EnKF yields more accurate and stable assimilation.
LSI enables joint learning of latent variables and generative models.
problem Joint optimization of latent variables and generative models.
method Developed a principled ELBO objective in continuous time for joint learning.
result LSI learns effective latent representations and generative transformations.
We develop a real-time anomaly detection algorithm for directed activity on large, sparse networks. We model the propensity for future activity using a dynamic logistic model with interaction terms for sender- and receiver-specific latent factors in addition to sender- and receiver-specific popularity scores; deviation…
Latent variable models are an elegant framework for capturing rich probabilistic dependencies in many applications. However, current approaches typically parametrize these models using conditional probability tables, and learning relies predominantly on local search heuristics such as Expectation Maximization. Using te…
An ADRC-incorporated SGD algorithm improves latent factor analysis speed and accuracy.
problem Slow convergence in standard SGD for HDI matrix analysis.
method Incorporates ADRC principles to refine historical and future learning error states.
result Empirically outperforms state-of-the-art LFA models in HDI matrix prediction.
Paper tackles causal effect identification in sub-population with latent variables.
problem Identify causal effects in a sub-population with latent variables.
method Extend relevant graphical definitions and propose an algorithm for the s-ID problem.
result Sound algorithm for s-ID problem with latent variables.
This work reduces the dimensionality of text data using SVD, improving performance and computational efficiency.
problem High-dimensional input spaces in text classification lead to excessive parameter count and computational infeasibility.
method Singular Value Decomposition (SVD) is applied to transform the input space into a lower-dimensional latent space.
result Neural networks trained on the lower-dimensional latent space achieve comparable or better performance with reduced computational complexity.