Bayesian neural networks with data augmentation show a persistent cold posterior effect.
problem Understanding the cold posterior effect in Bayesian neural networks with data augmentation.
method Developed principled Bayesian neural networks using data augmentation, providing exact likelihoods and tight bounds.
result The cold posterior effect persists even in models incorporating data augmentation, suggesting it's not an artifact.
The cold posterior effect is explored through PAC-Bayes bounds for small sample sizes.
problem The cold posterior effect in approximate Bayesian inference for small datasets.
method Investigation through PAC-Bayes generalization bounds, focusing on temperature parameter λ.
result The temperature parameter λ in PAC-Bayes bounds captures the cold posterior effect.
New research shows CPE only occurs when Bayesian posterior underfits.
problem Model misspecification leading to CPE under perfect model specification.
method Theoretical analysis of Bayesian posterior and underfitting.
result No CPE if there is no underfitting of the Bayesian posterior.
Cold posteriors improve Bayesian neural networks by reducing overestimation of aleatoric uncertainty.
problem Overestimation of aleatoric uncertainty in Bayesian neural networks.
method Tuning the temperature of the posterior on a validation set.
result Reducing temperature leads to better reflection of true prior beliefs.
Bayesian neural networks use temperature adjustments to improve predictive performance.
problem Lack of theoretical generalization guarantees for Bayesian neural networks.
method Temperature adjustments to balance likelihood and prior regularization.
result Improved predictive performance through temperature adjustments.
A new method selects optimal temperature for Bayesian Deep Learning.
problem Finding the optimal temperature for improving predictive performance in Bayesian Deep Learning.
method Data-driven approach to estimate temperature as a model parameter.
result Our method performs comparably to grid search but at a fraction of the cost.
Cold posteriors in BNNs harm performance, likely due to incorrect likelihood.
problem Cold posteriors in Bayesian neural networks degrade performance.
method Developed a generative model explaining cold posteriors and matched it to the tempered likelihoods.
result Cold posteriors are a result of using the wrong likelihood for image classification datasets.
Bayes posterior yields worse predictions than simpler methods in deep neural networks.
problem Understanding and improving predictive performance in Bayesian deep learning.
method Careful MCMC sampling and evaluation of cold posteriors.
result Cold posteriors yield significantly better predictive performance than the true Bayes posterior.
Bayesian priors for neural networks are improved by incorporating weight correlations and tail behavior.
problem Improving Bayesian priors for neural networks to better reflect true beliefs and performance.
method Analyzed summary statistics of neural network weights in different architectures and incorporated these observations into new priors.
result Improved performance on image classification datasets by using new priors that account for weight correlations and tail behavior.
New priors improve Bayesian neural networks without cooling.
problem Bayesian neural networks underfit on clean datasets.
method Introduce DirClip and confidence priors to replace cooling.
result DirClip and confidence priors outperform cold posterior.
Bayesian learning made scalable with posteriors library.
problem Computational challenges in Bayesian learning with modern models.
method Introducing posteriors library and tempered MCMC.
result Bayesian approximations are useful and scalable.
Bayesian classification improves with explicit aleatoric uncertainty.
problem Lack of aleatoric uncertainty representation in Bayesian classification.
method Explicitly account for aleatoric uncertainty using a Dirichlet observation model.
result Explicit aleatoric uncertainty improves performance of Bayesian neural networks.
BNNpriors library improves Bayesian neural network inference with various prior distributions.
problem Challenges in choosing good prior distributions for Bayesian neural networks.
method State-of-the-art Markov Chain Monte Carlo inference with a wide range of predefined priors.
result Facilitates foundational discoveries on the nature of the cold posterior effect.
High-dimensional unimodal distributions can cause MCMC methods to fail.
problem Failure of MCMC methods in high-dimensional unimodal distributions.
method Examples and theoretical analysis of MCMC methods, including Metropolis-Hastings adjusted methods.
result MCMC methods can take an exponential run-time for high-dimensional unimodal distributions.
MFVI can overestimate predictive variance compared to the exact posterior
problem MFVI underestimates posterior variance
method Analyzing conjugate Bayesian Linear Regression
result MFVI can overestimate predictive variance compared to the exact posterior
Bayesian neural networks show complex posterior distributions that HMC can capture effectively.
problem Understanding and approximating the high-dimensional, non-convex posterior of Bayesian neural networks.
method Full-batch Hamiltonian Monte Carlo (HMC) on modern architectures.
result HMC provides a robust and comparable representation of the BNN posterior, with significant performance gains over standard training and deep ensembles.
SPARC tackles cold-start nodes in graphs by using spectral embeddings.
problem Cold-start nodes in graphs lacking initial connections.
method Introduces SPARC, a framework utilizing spectral embeddings to predict on cold-start nodes.
result SPARC outperforms existing models on cold-start nodes across tasks.
Monotonic improvement in uncertainty estimation with Gaussian processes as dimension increases.
problem Uncertainty quantification in machine learning models, especially with Gaussian processes, is challenging and poorly understood.
method Analyzing the behavior of marginal likelihood and cross-validation metrics as input dimension increases, and exploring the effects of cold posteriors.
result The marginal likelihood improves monotonically with input dimension, while cross-validation metrics exhibit double descent behavior.
This work explains how tempering improves Bayesian neural networks by reducing the impact of data augmentation.
problem Improper sharpening of Bayesian neural networks leads to suboptimal performance.
method Theoretical analysis and empirical evaluations of simplified settings and group convolutions.
result Tempering reduces the misspecification due to modeling augmentations as independent and identically distributed (i.i.d.) data.
Restricted Boltzman Machines (RBMs) have been successfully used in recommender systems. However, as with most of other collaborative filtering techniques, it cannot solve cold start problems for there is no rating for a new item. In this paper, we first apply conditional RBM (CRBM) which could take extra information in…
Adaptive querying learns user psychometrics with AI personas.
problem Learning user psychometrics within query budgets.
method Persona-induced latent variable model with AI personas and large language model response distributions.
result Persona-based posteriors deliver accurate probabilistic predictions.
Study on cold and freezing sets in digital images.
problem Properties of cold sets in digital images.
method Analysis of properties and relationships between cold and freezing sets.
result Examined relationships between cold and freezing sets.
NetDP predicts loan defaults using network data, addressing cold-start issues.
problem Cold-start problem in default prediction for new users.
method Combines unsupervised and supervised network representations, using parameter-server for scalability.
result Effectiveness in cold-start problem, especially for new users.
Numerically estimates Colding-Minicozzi entropies of self-shrinkers.
problem Estimating Colding-Minicozzi entropies of self-shrinkers.
method Numerical estimation of entropies for specific self-shrinkers.
result Colding-Minicozzi entropies of n-dimensional Angenent torus are decreasing with dimension. Memory-Augmented Meta-Optimization improves cold-start recommendation.
problem Cold-start problem in recommender systems for new users or items.
method Memory-Augmented Meta-Optimization approach with personalized and task-specific memories.
result Significant improvement in cold-start recommendation performance on multiple datasets.
Study bounds self-shrinker entropy using Li-Yau volume and Colding-Minicozzi entropy.
problem Bounding entropy of self-shrinkers in arbitrary codimensions.
method Introduced stable conformal volume and virtual entropy to prove bounds.
result Entropy bounds are sharp and independent of codimension.
Playlist recommendation involves producing a set of songs that a user might enjoy. We investigate this problem in three cold-start scenarios: (i) cold playlists, where we recommend songs to form new personalised playlists for an existing user; (ii) cold users, where we recommend songs to form new playlists for a new us…
The paper investigates quantitative rigidity using Colding's monotonicity formulas for Ricci curvature.
problem Quantifying rigidity in manifolds with nonnegative Ricci curvature.
method Investigates pinching of Colding's monotone functionals and constructs k-splitting functions. result Quantitative control of splitting functions by pinching at independent points controls the distance to the nearest cone.
Graph neural networks improve cold start for new items in recommender systems.
problem Cold start problem for new items in recommender systems.
method Item hierarchy graphs and bespoke graph neural network architecture.
result Our method achieves better forecasting quality than state-of-the-art with comparable computational time.
The item cold-start problem seriously limits the recommendation performance of Collaborative Filtering (CF) methods when new items have either none or very little interactions. To solve this issue, many modern Internet applications propose to predict a new item's interaction from the possessing contents. However, it is…
Unified approach for conversational recommendation by integrating attributes and items.
problem Cold-start users' real-time personalization in online recommendation.
method Seamlessly unifies attributes and items in Thompson Sampling framework for interactive decision-making.
result Conversational Thompson Sampling (ConTS) outperforms existing methods in success rate and conversation turns.
Study shows rigidity for entropy minimizers in non-monotone cases.
problem Rigidity of entropy minimizers in non-monotone settings.
method Elementary proofs in non-monotone situations.
result Showed rigidity for minimizers of generalized Colding-Minicozzi entropies.
We address the cold start problem in recommendation systems assuming no contextual information is available neither about users, nor items. We consider the case in which we only have access to a set of ratings of items by users. Most of the existing works consider a batch setting, and use cross-validation to tune param…
This work explores the ability of collective matrix factorization models in recommender systems to make predictions about users and items for which there is side information available but no feedback or interactions data, and proposes a new formulation with a faster cold-start prediction formula that can be used in rea…
A new algorithm for better decision-making in recommendation systems.
problem Stochastic multi-armed bandit problem and cold start problem in recommender systems.
method Proposes Hellinger-UCB, a variant of UCB algorithm using squared Hellinger distance.
result Hellinger-UCB reaches the theoretical lower bound and outperforms other algorithms in practical applications.
Entropy derived from Colding's volume on Ricci-flat manifolds.
problem Deriving Perelman's entropy from Colding's monotonic volume.
method Applying Colding's monotonic volume to Perelman's N-space for harmonic functions on Ricci-flat manifolds.
result Entropy is the limit of Colding's monotonic volume.
Enhances time-series modeling by dropping patches, improving efficiency and adaptability.
problem Improving masked time-series modeling efficiency and adaptability.
method DropPatch method, which randomly drops sub-sequence level patches.
result DropPatch improves pre-training efficiency and adaptability in various learning scenarios.
We explain SSL objectives as log-likelihoods in a data curation model.
problem Lack of understanding of SSL objectives as log-likelihoods.
method Formulate SSL objectives as a log-likelihood in a generative model of data curation.
result SSL methods can be understood as lower-bounds on a principled log-likelihood.
FAB-COST improves cold-start recommendation accuracy with less data.
problem Cold-start problem in recommendation systems.
method Contextual bandit algorithm using Expectation Propagation and Assumed Density Filtering.
result FAB-COST outperforms Laplace approximation on real data.
NAM learns multiview representations with attention for better task performance.
problem Finding optimal combination of multiviews for specific tasks.
method Introduces Neural Attentive Multiview machine with attention mechanism.
result NAM outperforms single view and alternative multiview methods in item recommendations.
With the growing importance of personalized recommendation, numerous recommendation models have been proposed recently. Among them, Matrix Factorization (MF) based models are the most widely used in the recommendation field due to their high performance. However, MF based models suffer from cold start problems where us…
Proves effective linear volume growth for 3-manifolds with positive scalar curvature.
problem Volume growth of three-manifolds with positive scalar curvature.
method Utilizes the technique of μ-bubbles and almost-splitting theorem.
result Proves effective linear volume growth for 3-manifolds with non-negative Ricci curvature and uniformly positive scalar curvature.
Study minimal surfaces in complex hyperbolic space, linking entropy and volume.
problem Characterize minimal submanifolds in complex hyperbolic space.
method Analyze asymptotic regularity and introduce Colding-Minicozzi entropy and CR-volume.
result Establish a connection between Colding-Minicozzi entropy and CR-volume.
Motivated and inspired by the recent work of Colding [5] and Colding-Minicozzi [6] we derive several families of monotonicity formulas for manifolds with nonnegative Bakry-Emery Ricci curvature, extending the formulas in [5, 6].
Proof of Reifenberg theorem in metric spaces, expanding on Cheeger and Colding's work.
problem Proving the Reifenberg theorem in metric spaces using Gromov-Hausdorff distance.
method Detailed proof of Cheeger and Colding's result, expanding on their arguments.
result BiLipschitz version of the Reifenberg theorem in metric spaces.
In recommender systems, cold-start issues are situations where no previous events, e.g. ratings, are known for certain users or items. In this paper, we focus on the item cold-start problem. Both content information (e.g. item attributes) and initial user ratings are valuable for seizing users' preferences on a new ite…
Proposes DSM priors for Bayesian neural networks to improve interpretability and robustness.
problem Bayesian neural networks struggle with interpretability, overconfidence, and adversarial attacks.
method Introduces Dirichlet scale mixture (DSM) priors to address these issues.
result DSM priors lead to sparse networks, robustness against adversarial attacks, and competitive predictive performance.
Cold-start PV forecasting uses synthetic histories to train time-series foundation models.
problem Cold-start PV forecasting
method Zero-shot pipeline with synthetic histories
result TabPFN-TS achieves the lowest error under Real Feedback strategy