P3I learns holistic scene representations from a single image.
problem Inferring camera poses, object locations, and global scene structures from a single image.
method Combines search-based and gradient-based algorithms.
result P3I outperforms baselines on various image manipulation tasks.
Theoretical analysis shows LLMs can self-correct responses through in-context learning.
problem Understanding how large language models improve through self-correction.
method Theoretical analysis based on simplified alignment task, focusing on softmax attention, multi-head attention, and MLP blocks.
result LLMs can refine responses in an in-context way when given accurate self-examinations as rewards.
The paper corrects biases in estimating intrinsic dimension and differential entropy.
problem Systematic bias in estimating intrinsic dimension and differential entropy.
method A bias-corrected estimator for both measures is proposed, highlighting shared steps and useful consequences.
result Simultaneous estimation of differential entropy and intrinsic dimension provides complementary perspectives on underlying manifolds.
Let L be a nonunimodular definite lattice. Using a theorem of Elkies we show that whether L embeds in the standard definite lattice of the same rank is completely determined by a collection of lattice correction terms, one for each metabolizing subgroup of the discriminant group. As a topological application this gives…
New MCMC method corrects bias without extra cost.
problem Correcting bias in MCMC algorithms without additional computational cost.
method Generalized Markov Chain Importance Sampling methods.
result Proposed methods are more efficient than Metropolis-Hastings versions.
This paper improves deep learning model consistency through ensemble methods.
problem Consistency and correct-consistency issues in deep learning models.
method Formal definition of consistency and correct-consistency, proving ensemble improvement, proposing dynamic snapshot ensemble method.
result Ensemble methods can improve correct-consistency of deep learning models.
The paper develops methods to assess and correct model uncertainties in graphical models.
problem Model uncertainty in probabilistic graphical models.
method Information-theoretic and non-parametric stress tests.
result Ranking and correcting impactful sources of uncertainty in graphical models.
PCA whitening weighted by Zipfian word frequencies improves task performance.
problem Skewed word embedding spaces in neural models.
method PCA whitening weighted by empirical word frequencies following Zipf's law.
result Significantly improves task performance, surpassing baselines.
Mathematical models are used extensively for diverse tasks including analysis, optimization, and decision making. Frequently, those models are principled but imperfect representations of reality. This is either due to incomplete physical description of the underlying phenomenon (simplified governing equations, defectiv…
New CRB derived for curved models using extrinsic geometry.
problem Estimate curved statistical families accurately.
method Vector generalization of CRB with curvature correction using SDP and SOS relaxations.
result Directional curvature correction provides more accurate estimation.
Deep ensembles effectively capture epistemic uncertainty through training stochasticity, providing a frequentist perspective.
problem Understanding and quantifying epistemic uncertainty in machine learning models.
method Bootstrap-based estimator and decomposition of deep ensembles into data variability and training stochasticity.
result Deep ensembles primarily capture training stochasticity, explaining their effectiveness in quantifying epistemic uncertainty.
We present a numerical algorithm for nonnegative matrix factorization (NMF) problems under noisy separability. An NMF problem under separability can be stated as one of finding all vertices of the convex hull of data points. The research interest of this paper is to find the vectors as close to the vertices as possible…
Two approaches improve conformal Bayes for label shift, one post-hoc and one in-training.
problem Improving prediction sets for target domain under label shift.
method Two complementary approaches: post-hoc calibration and in-training adaptation.
result In-training adaptation achieves up to 43% width reduction at unchanged coverage.
Deep reinforcement learning has made significant progress in the field of continuous control, such as physical control and autonomous driving. However, it is challenging for a reinforcement model to learn a policy for each task sequentially due to catastrophic forgetting. Specifically, the model would forget knowledge …
Geometrically refines Cramér-Rao bound using extrinsic manifold curvature.
problem Improving estimator efficiency in non-asymptotic settings.
method Incorporates curvature-aware corrections based on extrinsic geometry of statistical model manifold.
result Meaningful tightening of estimator variance bounds.
Study 3d N=1 vacua from M-theory compactification on Spin(7) space.
problem Quantum corrections in 3d N=1 vacua from M-theory compactification.
method Use Higgs bundles to analyze 3d N=1 vacua and track corrections.
result Topological anomalies are robust and calculable in 3d effective field theory.
Paper proves noise-tolerant SSC using greedy methods under coherence conditions.
problem Proving noise-tolerant SSC using greedy methods under coherence conditions.
method Derives coherence-based sufficient conditions for correct neighbor identification using MP/OMP in the presence of bounded noise.
result MP/OMP succeed in identifying correct neighbors under certain noise levels, leading to higher clustering accuracy.
The paper addresses missing data imputation issues by correcting for distribution shift.
problem Missing data imputation and the resulting distribution shift between observed and full data.
method Formulates imputation as a risk minimization problem and proposes a novel algorithm to correct for distribution shift.
result The proposed algorithm consistently improves imputation accuracy, reducing RMSE and Wasserstein distance by 3% and 7%, respectively.
New theory challenges traditional machine learning assumptions.
problem Traditional machine learning theories are critiqued.
method A new theory is proposed and discussed.
result Learning true probabilities is not equivalent to other learning goals.
The separability assumption (Donoho & Stodden, 2003; Arora et al., 2012) turns non-negative matrix factorization (NMF) into a tractable problem. Recently, a new class of provably-correct NMF algorithms have emerged under this assumption. In this paper, we reformulate the separable NMF problem as that of finding the ext…
We study adversarial robustness of neural networks from a margin maximization perspective, where margins are defined as the distances from inputs to a classifier's decision boundary. Our study shows that maximizing margins can be achieved by minimizing the adversarial loss on the decision boundary at the "shortest succ…
Proposes EDM algorithm to accelerate model training in distributed networks.
problem Hindered effectiveness of distributed stochastic optimization algorithms due to data heterogeneity and network sparsity.
method Introduces Exact-Diffusion with Momentum (EDM) algorithm, incorporating momentum techniques to mitigate bias and enhance convergence rate.
result EDM algorithm converges sub-linearly to the optimal solution, radius independent of data heterogeneity, for non-convex objective functions.
Geometric framework explains and controls implicit bias in machine learning.
problem Understanding and controlling the selection of solutions in overparameterized models.
method Developed a theoretical and constructive framework based on geometric corrections induced by gradient noise and continuous symmetries of the loss.
result Computed the induced bias across various architectures and enabled inverse design to shape the bias.
A Bernoulli Mixture Model (BMM) is a finite mixture of random binary vectors with independent dimensions. The problem of clustering BMM data arises in a variety of real-world applications, ranging from population genetics to activity analysis in social networks. In this paper, we analyze the clusterability of BMMs from…
Study on how reparametrization affects neural nets' parameter spaces from a geometric perspective.
problem Inconsistencies in flatness measures, optimization, and probability densities under reparametrization.
method Riemannian geometry to study invariance of neural nets under reparametrization.
result Invariance of neural nets is an inherent property if the metric is explicitly represented and transformation rules are correct.
Framework improves target domain prediction using quantile matching.
problem Improving prediction accuracy in data-scarce target domains.
method Conditional quantile matching for distributional alignment.
result Empirical risk minimizer achieves tighter excess risk bound.
Semantic Web knowledge representation standards, and in particular RDF and OWL, often come endowed with a formal semantics which is considered to be of fundamental importance for the field. Reasoning, i.e., the drawing of logical inferences from knowledge expressed in such standards, is traditionally based on logical d…
Conformal Bayes under label shift: post-hoc calibration vs. in-training adaptation
problem Bayesian prediction sets under label shift
method Post-hoc calibration vs. In-training adaptation
result Both strategies achieve valid coverage equally in an unbiased training regime
Conformal prediction improves signal detection accuracy in railway images.
problem Improving the reliability of machine learning models for railway signal detection.
method Applying conformal prediction to a novel dataset of train operator perspective images.
result The approach enhances the reliability of machine learning models for detecting railway signals.
Gradient clipping helps private SGD converge despite potential bias.
problem Gradient clipping in private SGD can bias convergence.
method Theoretical analysis and empirical evaluation of gradient clipping effects.
result Gradient clipping can prevent convergence to stationary points and introduces bias.
Conditional forecasts of risk measures play an important role in internal risk management of financial institutions as well as in regulatory capital calculations. In order to assess forecasting performance of a risk measurement procedure, risk measure forecasts are compared to the realized financial losses over a perio…
This paper shows that pairwise PageRank orders emerge from two-hop walks. The main tool used here refers to a specially designed sign-mirror function and a parameter curve, whose low-order derivative information implies pairwise PageRank orders with high probability. We study the pairwise correct rate by placing the Go…
New method for community detection in graphs faster than DCBM inference.
problem Efficiently detecting communities in graphs with heterogeneous node degrees.
method Reformulated constrained nonnegative matrix factorization for DCBM inference.
result Faster community detection (4 minutes for 100k nodes vs. DCBM's 10+ minutes).
ProPublica's analysis of recidivism predictions produced by Correctional Offender Management Profiling for Alternative Sanctions (COMPAS) software tool for the task, has shown that the predictions were racially biased against African American defendants. We analyze the COMPAS data using a causal reformulation of the un…
Study investigates how simple speech sounds can form abstract categories.
problem How do abstract categories like phonemes emerge from speech exposure?
method Used modeling techniques to test Memory-Based Learning and Error-Correction Learning.
result Error-Correction Learning models can learn abstractions, identifying phone inventory and grouping.
Semiparametric method removes bias in functional bilevel gradient estimation.
problem First-order bias in plug-in hypergradient when lower-level problem is nonparametric.
method Semiparametric debiasing theory based on efficient influence function leads to cross-fitted orthogonal hypergradient estimator.
result Asymptotic normality and uniform control over outer parameter established for the estimator.
New algorithm radVI improves variational inference by optimizing radial profiles.
problem Gaussian approximations often fail to capture the radial profile of complex distributions.
method Optimizes over radial profiles in variational inference, providing theoretical guarantees.
result Theoretical convergence guarantees for radVI, improving over existing VI methods.
In this paper we establish a constructive method in order to show global existence and regularity for a class of degenerate parabolic Cauchy problems which satisfy a weak Hoermander condition on a subset of the domain where the data are measurable and which have regular data on the complementary set of the domain. This…
Generating and eliminating adversarial examples has been an intriguing topic in the field of deep learning. While previous research verified that adversarial attacks are often fragile and can be defended via image-level processing, it remains unclear how high-level features are perturbed by such attacks. We investigate…
The paper explores how LLMs with CoT improve performance on complex tasks.
problem Understanding the mechanisms behind LLMs' improved performance with CoT.
method Using circuit complexity theory, the paper examines LLMs' expressivity in solving mathematical and decision-making problems.
result LLMs with CoT can generate correct solutions step-by-step, even for complex tasks.
The paper analyzes DeepWalk and node2vec for community detection in stochastic blockmodels.
problem Theoretical understanding of community detection in stochastic blockmodels.
method Matrix factorization perspective on DeepWalk and node2vec.
result Node2vec and DeepWalk can perfectly recover community memberships in sparse graphs.
Estimates citation impact to recommend best publication venue.
problem Choosing optimal publication venue for academic papers.
method Treatment effect estimation and bias correction method.
result Effective recommendation of publication venues based on citation potential.
Unified framework for comparing clusterings from information-theoretic and pair-counting perspectives.
problem Divergent evaluations of unsupervised models due to different clustering similarity measures.
method Developed an analytical framework that unifies pair-counting and information-theoretic clustering similarity measures.
result Unified framework clarifies when and why the two regimes diverge and provides a principled basis for selecting and interpreting clustering similarity measures.
In many learning problems, the training and testing data follow different distributions and a particularly common situation is the \textit{covariate shift}. To correct for sampling biases, most approaches, including the popular kernel mean matching (KMM), focus on estimating the importance weights between the two distr…
We describe an exercise of using Big Data to predict the Michigan Consumer Sentiment Index, a widely used indicator of the state of confidence in the US economy. We carry out the exercise from a pure ex ante perspective. We use the methodology of algorithmic text analysis of an archive of brokers' reports over the peri…
Non-intrusive algorithm corrects corrupted data in classification models.
problem Classification problems with corrupted training data.
method Post-processes a trained classification model with a correction procedure.
result Corrected models deliver correct classification results even with corrupted data.
A large number of objectives have been proposed to train latent variable generative models. We show that many of them are Lagrangian dual functions of the same primal optimization problem. The primal problem optimizes the mutual information between latent and visible variables, subject to the constraints of accurately …
Motivated by an important insight from neural science, we propose a new framework for understanding the success of the recently proposed "maxout" networks. The framework is based on encoding information on sparse pathways and recognizing the correct pathway at inference time. Elaborating further on this insight, we pro…