Bayesian models evaluate sentence comprehension, showing direct access model fits data better.
problem Evaluating models of retrieval in sentence comprehension.
method Implemented Bayesian hierarchical models to compare activation-based and direct access models.
result Direct access model fits data better than activation-based model.
Bayesian model explains sentence comprehension better than distance-based account.
problem Sentence comprehension latency and dependency distance.
method Bayesian hierarchical mixture process modeling.
result Direct-access model fits Chinese relative clause data better.
The paper proposes a method to learn Bayesian networks with low rank conditional probability tables.
problem Learning the structure of Bayesian networks efficiently.
method Introduces low rankness for conditional probability tables, connects to Fourier transformation, and proposes a polynomial time algorithm.
result Correctly recovers the true directed structure of a low rank Bayesian network with few queries and polynomial samples.
A proof that hyperbolic plane cannot be immersed in Euclidean 3-space.
problem Proving the impossibility of isometrically immersing the hyperbolic plane in Euclidean 3-space.
method Applying ideas from undergraduate mathematics, including moving frames and connection forms, to simplify the proof.
result A key transition from principal directions to asymptotic directions simplifies the proof and yields a coordinate system.
Generative models learn from biased data using weighted importance.
problem Learning from biased or related data distributions.
method Importance weighting to estimate loss with respect to target distribution.
result Effective in various settings with theoretical guarantees and good performance.
Framework reuses pre-trained models for data-free transfer learning.
problem Challenges in retrieving source data for model training.
method Model Recycling Framework for parameter-efficient training.
result Makes multi-source data-free supervised transfer learning possible.
Paper develops methods for fair insurance pricing without direct access to sensitive attributes.
problem Fairness in insurance pricing with restricted access to sensitive attributes.
method Develops statistical methods for estimating discrimination-free premiums using privatized sensitive attributes.
result The proposed methods enable fair insurance pricing while respecting privacy and regulatory constraints.
We consider the partial observability model for multi-armed bandits, introduced by Mannor and Shamir. Our main result is a characterization of regret in the directed observability model in terms of the dominating and independence numbers of the observability graph. We also show that in the undirected case, the learner …
Method differentiates diffusion model training to predict sample sensitivity.
problem Predict how diffusion model samples change with small perturbations.
method Closed-form procedure for computing directional derivatives of the map.
result Estimates sensitivity of diffusion model samples to additive perturbations.
Query access significantly speeds up learning Multi-Index Models under Gaussian distribution.
problem Agnostically learning Multi-Index Models (MIMs) under Gaussian distribution.
method Query access for MIMs with complexity O(k)poly(1/ε)poly(d) under standard regularity assumptions. result Query access gives significant runtime improvements over random examples for agnostically learning MIMs.
New algorithm extracts features from superpositions in machine learning models.
problem Challenges in extracting interpretable features from complex models in superposition.
method An efficient query algorithm that identifies non-degenerate feature directions and reconstructs the function.
result Identifies all feature directions whose responses are non-degenerate and reconstructs the function \( f \) in a general superposition setting.
SPIBB improves safe policy training without direct access to the environment.
problem Training safe policies without direct access to the environment.
method SPIBB with Baseline Bootstrapping, inspired by knows-what-it-knows paradigm.
result SPIBB improves safety and mean performance compared to existing algorithms.
This work formalizes and solves GNN extraction, a new type of attack.
problem Extracting the underlying model from a black-box GNN with API access.
method Formalizes GNN extraction, presents a solution with preliminary results.
result Preliminary results show the feasibility of GNN extraction.
WAFFLE embeds watermarks in federated learning models without access to training data.
problem Model theft in federated learning due to lack of centralized control.
method Retraining step at server after model aggregation, generating backdoor as watermark.
result Efficient embedding of resilient watermark with negligible accuracy loss.
Develops Frank-Wolfe Augmented Lagrangian for convex optimization.
problem Minimizing functions over intersections of convex sets.
method Frank-Wolfe Augmented Lagrangian (FW-AL) method.
result Sublinear convergence rate for general convex compact sets, linear for polytopes.
We propose a unified framework in which the different constructions of cohomology groups for topological and Lie groups can all be treated on equal footings. In particular, we show that the cohomology of "locally continuous" cochains (respectively "locally smooth" in the case of Lie groups) fits into this framework, wh…
Generative Distributionally Robust Optimization (GDRO) improves model compatibility and adversarial structure in DRO.
problem Trade-off between model compatibility and adversarial structure in existing DRO methods.
method GDRO accepts any sampleable conditional generator and restricts worst-case laws to a chosen family, using sampler-Sinkhorn pairing.
result Reduces inventory regret by 60% and navigation collisions by 50% relative to nominal decisions.
Private method measures nonlinear correlations between data hosted across two entities.
problem Measuring nonlinear correlations between sensitive data hosted across multiple parties while preserving privacy.
method Differentially private estimator of distance correlation.
result First private estimator of nonlinear correlations in a multi-party setup.
Survey of methods for learning from observation without requiring expert actions.
problem Lack of access to expert actions in imitation learning.
method Survey and classification of state-only imitation learning (SOIL) methods.
result Identification of open problems and future research directions.
Optimal domain adaptation model using Fisher's Linear Discriminant.
problem Improving classification accuracy across different domains.
method Convex combination of source and target hypotheses, derived under 0-1 loss.
result Effective classifier can be computed without direct source task information.
Federated learning improves keyboard suggestions without user data.
problem Improving search suggestion quality without direct user data access.
method Applied federated learning for global-scale model training and deployment.
result Significant quality increases in search suggestions.
The study examines discrete states in hyperbolic spaces using specific transformations.
problem Characterizing discrete states in hyperbolic spaces via specific transformations.
method Analyzing two-parameter families of subgroups in hyperbolic planes and spaces with up to four generators.
result Discreteness of accessible states in hyperbolic spaces is determined for specific transformations.
Gradient-free method solves infinite-dimensional optimization problems.
problem Optimizing functions in infinite-dimensional spaces.
method Uses directional derivatives and a pre-basis for Hilbert space.
result Proves convergence for solving PDEs using PINNs.
New framework PBBO optimizes latent functions with preferential feedback.
problem Optimizing latent functions with parallel preferential feedback.
method Gaussian process model with a likelihood for parallel comparisons.
result Framework PBBO generalizes and expands Bayesian optimization.
New framework improves offline RL performance without complex methods.
problem Limited access to online interactions in real-world RL applications.
method Behavior Regularized Actor Critic (BRAC) framework.
result Many recent technical complexities are not necessary for strong offline RL performance.
Quantum algorithm finds extremal values without direct function access.
problem Finding extremal values of hidden functions without direct access.
method Parametric quantum circuit trained with a trainable quantum feature map.
result Algorithm successfully finds extremal values even with sparse training data.
We introduce a new measure of activity of financial markets that provides a direct access to their level of endogeneity. This measure quantifies how much of price changes are due to endogenous feedback processes, as opposed to exogenous news. For this, we calibrate the self-excited conditional Poisson Hawkes model, whi…
A diffusion model estimates data manifold dimension by tracking likelihood increases.
problem Estimating the intrinsic dimension of data manifolds.
method Trained diffusion model approximates score function, revealing manifold directionality.
result Diffusion model provides an approximation of the tangent space's dimension.
We present a new approach for the pricing of interest rate derivatives which allows a direct computation of option premiums without deriving a (Black-Scholes type) partial differential equation and without explicitly solving the stochastic process for the underlying variable. The approach is tested by rederiving the pr…
Random forest uses only triplet comparisons to learn from metric spaces.
problem Learning from metric spaces without direct access to data or distances.
method A novel random forest algorithm that uses only triplet comparisons.
result The proposed random forest is consistent and competitive with other methods.
GCPN uses reinforcement learning to generate molecules optimizing desired properties.
problem Generating novel molecules with desired properties while obeying physical laws.
method Graph Convolutional Policy Network (GCPN) trained with reinforcement learning.
result GCPN achieves significant improvements in molecule optimization tasks.
This paper uses deep learning to improve memory prefetching.
problem Improving memory performance in software optimizations.
method Relating prefetching strategies to n-gram models and using recurrent neural networks.
result Neural networks consistently outperform traditional prefetching strategies in benchmark tests.
A new method optimizes neural sequence models for better task performance.
problem Training neural sequence models with maximum likelihood estimation ignores task losses.
method Maximum likelihood guided parameter search (MGS) in the parameter space.
result MGS optimizes sequence-level losses, reducing repetition and non-termination.
Quantum neural networks need both data-dependent and trainable unitaries for effective geometric deformation.
problem Quantum neural networks lack the geometric flexibility of classical networks due to limitations in state reachability.
method Viewing quantum states as embedded manifolds, we analyze infinitesimal unitary actions and introduce the CLA maps and aCLS criterion.
result Geometric flexibility in quantum neural networks requires a joint dependence on data and trainable weights.
New model uncovers non-Euclidean neural representations.
problem Discovering latent neural states in complex, non-Euclidean spaces.
method Manifold GPLVM for identifying latent variables and neural contributions.
result mGPLVM correctly recovers non-Euclidean latent structures in neural data.
We address a fundamental problem that is systematically encountered when modeling complex systems: the limitedness of the information available. In the case of economic and financial networks, privacy issues severely limit the information that can be accessed and, as a consequence, the possibility of correctly estimati…
New method uses latent variables to estimate treatment effects from single-arm trials.
problem Estimating treatment effects from single-arm trials due to lack of external control groups.
method Latent-variable modeling with amortized variational inference for patient matching and direct effect estimation.
result Improved performance in direct treatment effect estimation and effect estimation via patient matching compared to previous methods.
MixDiff detects OOD samples in constrained access environments by comparing perturbed samples.
problem Detecting out-of-distribution samples in models with restricted access.
method Apply identical perturbation to target and similar ID sample, compare model outputs.
result MixDiff enhances OOD detection performance consistently across various datasets.
Prototype for adaptive electron microscopy scans reduces dose and time.
problem Reduce electron microscopy scan time and dose with minimal loss.
method Adaptive partial scanning with reinforcement learning.
result Reinforcement learning trained neural network optimizes scan paths.
New IPC data set for graph learning tasks.
problem Benchmarking graph-based machine learning methods.
method Compilation from International Planning Competitions (IPC).
result Distinctly different characteristics from popular benchmarks.
New algorithm infers reward function from agent's learning trajectories.
problem Inferring reward function from agent's learning data.
method Gradient-based approach to recover reward function.
result Improved performance compared to state-of-the-art methods.
Paper distinguishes causal structures under latent confounding and selection bias.
problem Distinguishing causal relationships when latent variables and selection bias are present.
method Formulated selected-marginalized directed graphs (smDGs) to distinguish causal structures.
result Two causal structures are indistinguishable if they have the same selected-marginalized directed graph.
Role mining tackles the problem of finding a role-based access control (RBAC) configuration, given an access-control matrix assigning users to access permissions as input. Most role mining approaches work by constructing a large set of candidate roles and use a greedy selection strategy to iteratively pick a small subs…
PBO optimizes functions with duels, reducing comparisons needed.
problem Optimizing functions with limited direct queries and user preferences.
method PBO uses Gaussian process models with Bernoulli likelihood to model preferences and define acquisition functions.
result PBO finds the optimum with fewer comparisons than previous methods.
This book is expository and is in Russian (sample English translation of two pages is given). It is shown how in the course of solution of interesting geometric problems (close to applications) naturally appear different notions of curvature, which distinguish given geometry from the 'ordinary' one. Direct elementary d…
New method lures adversaries to choose false directions in black-box attacks.
problem Improving robustness against black-box transfer attacks.
method Incorporates a removable neural network to induce a luring effect.
result Demonstrates effectiveness in thwarting state-of-the-art attacks.
Estimates TV distance between autoregressive models under different access models.
problem Estimating the total variation distance between two autoregressive distributions.
method Three access models: sample access, logit access, and noisy logit access; provides query complexity for each.
result Improved query complexity for estimating TV distance in autoregressive models.
We analyze the Hessian spectra of large models up to 100B parameters.
problem Accurate Hessian spectra of large foundation models are difficult to obtain.
method We use shard-local finite-difference Hessian vector products and stochastic Lanczos quadrature.
result We produce the first large-scale spectral density estimates of foundation models.