The paper proposes a method to stabilize predictions by identifying causal variables using a seed variable.
problem Stable prediction across unknown test data with potential spurious correlations.
method Conditional independence test based algorithm using a seed variable to separate causal from non-causal variables.
result The algorithm precisely separates causal and non-causal variables for stable prediction across test data.
Stable health predictions need deconfounding test set features.
problem Stability of predictions in health machine learning is compromised by selection biases.
method Deconfounding the test set features improves prediction stability across different environments.
result Improved stability achieved by deconfounding test set features.
I-SPEC learns stable models from data without full causal knowledge.
problem Learning models that generalize well across shifts in environment.
method End-to-end framework using partial ancestral graph to learn stable interventional distribution.
result I-SPEC can learn robust models without full causal knowledge.
SFB uses stable features to adapt unstable ones for better performance.
problem Improving classifier performance on out-of-distribution data by leveraging stable features.
method SFB learns a predictor that separates stable and unstable features, then adapts unstable predictions using stable predictions.
result SFB can learn an asymptotically-optimal predictor without test-domain labels.
The paper addresses online prediction in marginally stable systems with bounded perturbations.
problem Online prediction in marginally stable linear dynamical systems with adversarial or stochastic perturbations.
method The online least-squares algorithm is used to achieve sublinear regret, with a refined regret analysis and a structural lemma.
result The online least-squares algorithm achieves sublinear regret, with polynomial dependence on the system's parameters.
Retraining stabilizes model influence on data.
problem Performativity in predictive models leads to feedback loops.
method Developed the stable signal principle to address retraining dynamics.
result Repeated risk minimization converges geometrically to stable signal direction.
Proposes BSSP to stabilize predictions in biased data.
problem Distribution shift between training and test data causes prediction instability.
method Balance-subsampled stable prediction (BSSP) algorithm based on fractional factorial design.
result Significantly improves prediction stability across unknown test data.
In many important machine learning applications, the training distribution used to learn a probabilistic classifier differs from the testing distribution on which the classifier will be used to make predictions. Traditional methods correct the distribution shift by reweighting the training data with the ratio of the de…
Model predicts stable molecules with AI and physics constraints.
problem Designing stable molecules with limited data.
method Graph Scattering Variational Autoencoder with physical constraints.
result Model generates stable molecules with desired properties.
Stable long-term predictions for fluid flows using neural networks.
problem Predicting complex dynamics of fluid flows with high temporal stability.
method End-to-end trained neural network architecture combining CNN for spatial compression and LSTM for temporal prediction.
result Novel latent space subdivision (LSS) allows stable and controllable long-term predictions.
New algorithms improve policy evaluation in reinforcement learning.
problem Off-policy stability and on-policy efficiency issues in policy evaluation.
method Introduced novel algorithms using oblique projection method.
result Demonstrated both off-policy stability and on-policy efficiency.
Proposes a method to learn stable invariant sets in dynamical systems.
problem Learning stable invariant sets in general dynamical systems.
method Generalizes Manek and Kolter's approach by introducing projection onto latent space shapes and using invertible neural networks.
result Validates the method and shows its usefulness for long-term prediction.
We study binary classification algorithms for which the prediction on any point is not too sensitive to individual examples in the dataset. Specifically, we consider the notions of uniform stability (Bousquet and Elisseeff, 2001) and prediction privacy (Dwork and Feldman, 2018). Previous work on these notions shows how…
Successful attempts to predict judges' votes shed light into how legal decisions are made and, ultimately, into the behavior and evolution of the judiciary. Here, we investigate to what extent it is possible to make predictions of a justice's vote based on the other justices' votes in the same case. For our predictions…
Spectral gradient methods outperform Euclidean in certain deep learning scenarios.
problem When do spectral gradient updates outperform Euclidean in deep learning?
method Layerwise condition comparing squared nuclear-to-Frobenius ratio to stable rank of activations.
result Spectral updates can be more effective than Euclidean in deep networks and transformers.
LOO-StabCP speeds up CP for multiple predictions.
problem Balancing computational efficiency and prediction accuracy in CP.
method Leave-One-Out Stable Conformal Prediction (LOO-StabCP) using algorithmic stability.
result LOO-StabCP is faster and more accurate than RO-StabCP.
State-of-the-art learning algorithms, such as random forests or neural networks, are often qualified as "black-boxes" because of the high number and complexity of operations involved in their prediction mechanism. This lack of interpretability is a strong limitation for applications involving critical decisions, typica…
A new Bayesian approach to linear system identification has been proposed in a series of recent papers. The main idea is to frame linear system identification as predictor estimation in an infinite dimensional space, with the aid of regularization/Bayesian techniques. This approach guarantees the identification of stab…
Method predicts LFSM increments from past observations using codifference.
problem Forecasting LFSM increments from discrete-time observations.
method Uses codifference for serial dependence, with conditional expectation or projection for α>1 or α<2. result Method shows promising performance in forecasting volatilities, capturing kurtosis and serial dependence.
Bayesian Invariant Prediction models stable features from multi-environment data.
problem Analyzing stable features across multiple environments for better prediction and understanding.
method Developed Bayesian Invariant Prediction (BIP) model that encodes invariant feature indices as latent variables and infers them via posterior inference.
result BIP and its variational approximation (VI-BIP) outperform existing methods in accuracy and scalability for invariant prediction.
Collaborative filtering (CF) is a popular technique in today's recommender systems, and matrix approximation-based CF methods have achieved great success in both rating prediction and top-N recommendation tasks. However, real-world user-item rating matrices are typically sparse, incomplete and noisy, which introduce ch…
Stable deep models learn dynamical systems with formal stability guarantees.
problem Difficulties in making formal claims about stability of deep network dynamics models.
method Jointly learning a dynamics model and Lyapunov function to ensure non-expansiveness.
result Proposes an approach for stable deep learning of dynamical systems.
Study compares MoE and RNN models for stock price prediction across volatility profiles.
problem Improving stock price prediction accuracy across different volatility levels.
method Dynamic Mixture of Experts model combining RNN and linear models, adjusting weights through a gating network.
result MoE model outperforms individual models in reducing prediction errors.
The CFR framework has been a powerful tool for solving large-scale extensive-form games in practice. However, the theoretical rate at which past CFR-based algorithms converge to the Nash equilibrium is on the order of O(T−1/2), where T is the number of iterations. In contrast, first-order methods can be used to …
A cubing strategy identifies stable hyperparameter regions for uncertainty quantification in spatial deep learning.
problem Uncertainty quantification in spatial deep learning models.
method Cubing-based diagnostic framework to recursively partition hyperparameter space and evaluate regions using scoring rules.
result Our approach produces competitive or superior predictive intervals compared to a statistical baseline model.
Predictive coding networks are shown to be stable, robust, and converge faster than backpropagation.
problem Stability, robustness, and convergence of predictive coding networks.
method Dynamical systems theory and Lyapunov stability analysis.
result Predictive coding networks are Lyapunov stable and converge faster than backpropagation.
Learning kinetic systems from data is one of the core challenges in many fields. Identifying stable models is essential for the generalization capabilities of data-driven inference. We introduce a computationally efficient framework, called CausalKinetiX, that identifies structure from discrete time, noisy observations…
FIRES framework selects stable features from online data.
problem Efficiently selecting features in online settings with limited data.
method FIRES framework uses model parameter importance for feature selection.
result FIRES selects stable feature sets with minimal model complexity.
SIRUS creates interpretable rules from random forests for regression.
problem Lack of interpretability in complex machine learning models.
method Random forest with rule extraction for stability and simplicity.
result SIRUS produces stable and interpretable rule sets.
Study homotopy types of free racks and quandles, proving analogs of Milnor's theorem.
problem Understanding the homotopy types of free racks and quandles.
method Proved analogs of Milnor's theorem for racks and quandles and their pointed variants.
result Identified the homotopy types of free racks and quandles on spaces of generators.
The excited states of polyatomic systems are rather complex, and often exhibit meta-stable dynamical behaviors. Static analysis of reaction pathway often fails to sufficiently characterize excited state motions due to their highly non-equilibrium nature. Here, we proposed a time series guided clustering algorithm to ge…
Recent interest in the external validity of prediction models (i.e., the problem of different train and test distributions, known as dataset shift) has produced many methods for finding predictive distributions that are invariant to dataset shifts and can be used for prediction in new, unseen environments. However, the…
Online algorithms stabilize in feedback loops of performative prediction.
problem Feedback loops in algorithmic predictions influence data distributions.
method Martingale argument and randomization to avoid distributional assumptions.
result No-regret algorithms converge to performatively stable equilibria.
Random forests are stable and provide reliable prediction intervals.
problem Stability and reliability of random forest prediction intervals.
method Established stability under mild conditions and proved coverage bounds.
result Non-asymptotic lower and upper bounds for prediction interval coverage.
We consider the problem of learning linear prediction models with model misspecification bias. In such case, the collinearity among input variables may inflate the error of parameter estimation, resulting in instability of prediction results when training and test distributions do not match. In this paper we theoretica…
This paper optimizes performative risk by focusing on convex properties and developing efficient algorithms.
problem Performative risk, the loss experienced by decision makers, is not optimized by stable models.
method Identifying convex properties of loss function and model-induced distribution shift, developing algorithms for optimization.
result Optimization of performative risk with better sample efficiency than generic methods.
This paper introduces libconform v0.1.0, a Python library for the conformal prediction framework, licensed under the MIT-license. libconform is not yet stable. This paper describes the main algorithms implemented and documents the API of libconform. Also some details about the implementation and changes in future versi…
Algorithm learns dynamics from past observations.
problem Learning a nonlinear dynamical system.
method Spectral filtering, online convex optimization.
result Vanishing prediction error for marginally stable systems.
LSTM model predicts stock prices with high accuracy in stable sectors but struggles with volatile ones.
problem Predicting stock prices in emerging markets with limited data.
method Developed and evaluated an LSTM network on historical OHLCV data and technical indicators.
result Strong predictive performance (R2>0.87) for stable sectors, but challenges for volatile ones. FlowMM models stable crystal structures efficiently.
problem Predicting and proposing stable crystalline structures.
method Riemannian Flow Matching generalized to crystal symmetries.
result 3x more efficient at finding stable materials.
We present a new algorithm to generate minimal, stable, and symbolic corrections to an input that will cause a neural network with ReLU activations to change its output. We argue that such a correction is a useful way to provide feedback to a user when the network's output is different from a desired output. Our algori…
Study improves stock price prediction using adaptive Mixture of Experts framework.
problem Tackles diverse volatility regimes in stock price prediction.
method Combines RNN for high-volatility stocks and linear regression for stable stocks with a gating mechanism.
result Achieves up to 33% improvement in MSE for volatile assets and 28% for stable assets.
Thermalizer stabilizes autoregressive models for long-term predictions in chaotic systems.
problem Long-term predictions in chaotic spatiotemporal systems are unreliable due to trajectory divergence.
method Diffusion models are used to implicitly estimate the score of an invariant measure, which stabilizes autoregressive emulators by applying denoising during inference.
result Thermalization extends the time horizon of stable predictions by an order of magnitude in chaotic systems.
Improved online prediction with guaranteed coverage.
problem Creating reliable online predictions for arbitrary sequences.
method Online conformal prediction with decaying step sizes.
result Substantially improved practical properties, including close coverage at every time point.
Deep learning reveals lagged correlations in stock markets, showing accuracy decreases with shorter prediction horizons.
problem Capturing non-linear interactions in financial prediction problems using large-scale datasets.
method Applying deep learning to econometrically constructed gradients to learn and exploit lagged correlations among S&P 500 stocks.
result Model accuracies decrease with shorter prediction horizons, but remain significant in both stable and volatile markets.
New algorithm learns stable LDSs with lower error and better control performance.
problem Learning stable LDSs from data with minimal reconstruction error and stability constraints.
method Proposes an optimization method using a recent characterization of stable matrices, iteratively improving reconstruction error and ensuring stability.
result Achieves orders-of-magnitude improvement in reconstruction error compared to existing methods.
A framework uses preprocessing to improve psychiatric questionnaire predictions while maintaining interpretability.
problem Weak predictive accuracy and limited interpretability of psychiatric questionnaires.
method Two-stage method: stable preprocessing followed by a linear mapping.
result REFINE outperforms other interpretable approaches in psychiatric and non-psychiatric prediction tasks.
Model-based collaborative filtering analyzes user-item interactions to infer latent factors that represent user preferences and item characteristics in order to predict future interactions. Most collaborative filtering algorithms assume that these latent factors are static, although it has been shown that user preferen…