Foundation models improve on econometric benchmarks for forecasting volatility, but vary widely across models.
problem Comparing pretrained time series foundation models to econometric benchmarks for volatility forecasting.
method Systematic comparison of nine zero-shot TSFMs against eight econometric specifications on 50 assets across 3 markets and 3 horizons.
result Tiny Time Mixers (TTM) is the only model that consistently beats the Log-HAR benchmark, but performance varies widely across models.
EAGLE-Net enhances foundation models by integrating patch-level features for better tissue understanding.
problem Foundation models lack mechanisms for global tissue structure and local context in computational pathology.
method EAGLE-Net combines multi-scale spatial encoding, attention-guided loss functions, and background suppression to aggregate patch-level features into slide-level predictions.
result EAGLE-Net improves classification accuracy and concordance indices across multiple cancer types, producing biologically coherent attention maps.
Survey of financial foundation models for diverse applications.
problem Challenges in applying general-purpose FMs to financial tasks.
method Review of financial foundation models (FFMs) in three modalities.
result Emergence of FFMs designed specifically for finance.
PFNs pre-train models on simulated data to predict class probabilities.
problem Training machine learning models on large datasets.
method Pre-train a fixed model on small simulated datasets and use it to infer class probabilities in-context.
result PFNs achieve state-of-the-art performance and improve with larger inference data.
ProbFM provides principled uncertainty quantification for financial forecasting.
problem Lack of principled uncertainty quantification in financial applications.
method Probabilistic Time Series Foundation Model with Uncertainty Decomposition using Deep Evidential Regression (DER).
result DER maintains competitive forecasting accuracy while providing explicit epistemic-aleatoric uncertainty decomposition.
New definition of interpretability makes model design more actionable.
problem Current definitions of interpretability are not actionable and inform users poorly.
method Proposes a new definition of interpretability that is general, simple, and actionable.
result New definition reveals necessary properties for designing interpretable models.
Study evaluates interpretability of time series foundation models' latent spaces.
problem Improving interpretability of latent spaces in time series models for visual analytics.
method Evaluated MOMENT family of transformer-based models on five datasets, fine-tuning for performance.
result Fine-tuning improved latent space clarity but limited interpretability remained.
Foundation models fail to preserve continuous geometry, identified as the Geometric Alignment Tax.
problem Continuous geometry is lost in foundation models due to discrete categorical bottlenecks.
method Controlled ablations on synthetic systems and evaluation of 14 biological models using rate-distortion theory and MINE.
result Replacing cross-entropy with a continuous head reduces geometric distortion by up to 8.5x.
TSFMs show redundant components in layers, affecting their performance.
problem Redundant components in TSFMs layers impact their predictive accuracy.
method Mechanistic interpretability tools, ablations, logit attribution, theoretical framework.
result TSFMs are robust to ablations of entire layers and specific heads.
A new method calibrates scientific models by adding randomness to their predictions.
problem Current scientific foundation models lack calibrated uncertainty.
method Stochastic Attention, which randomizes attention weights using multinomial samples.
result Stochastic Attention achieves the strongest native calibration and sharpest prediction intervals.
InfoAtlas speeds up MI estimation for real-time data analysis.
problem Efficiently measuring statistical dependency between high-dimensional datasets.
method Directly infers mutual information in a single forward pass using a pretrained model.
result Matches state-of-the-art accuracy with 100x speedup.
CInA method uses attention to improve causal inference.
problem Challenges in causal inference, especially in complex tasks.
method CInA method utilizes self-supervised causal learning with multiple unlabeled datasets and transformer-type architecture.
result CInA effectively generalizes to out-of-distribution datasets and various real-world datasets.
New graph foundation models respect symmetries for broader applicability.
problem Tailored graph machine learning architectures limit broader applicability.
method Investigates symmetries for label and feature permutations, proving network universal approximator.
result Universal approximator on multisets respecting node and feature permutations.
New findings on how convolutional architectures approximate time series data.
problem Understanding the approximation properties of convolutional architectures in time series modeling.
method Mathematical analysis of convolutional architectures applied to time series modeling.
result A new definition of spectrum-based regularity for measuring temporal relationships under convolutional approximation.
The paper proposes a new risk model for foundation models in finance.
problem Understanding how foundation models affect trading strategies' risk and return.
method An extension of the CAPM, separating systematic and idiosyncratic risks.
result Monte Carlo dropout measures the epistemic risk of foundation models.
Single model learns physics from diverse data.
problem Lack of universal physics models for diverse applications.
method General Physics Transformer (GPhyT) trained on diverse physics data.
result Single model achieves superior performance across multiple physics domains.
FinRL-X unifies trading components for AI and rule-based strategies.
problem Inconsistent between research and live deployment in trading platforms.
method Modular architecture integrating data processing, strategy construction, backtesting, and execution.
result Unified protocol supports AI and rule-based trading components without altering execution.
Energy Transformer integrates attention, energy models, and associative memory.
problem Lack of clear theoretical foundations in attention mechanisms and straightforward design of energy functions in energy-based models.
method Proposes Energy Transformer, a sequence of attention layers with a specifically engineered energy function.
result Obtained strong results on graph anomaly detection and classification tasks.
Framework for training-free guidance in discrete diffusion models for molecular generation.
problem No equivalent training-free guidance methods for discrete diffusion models.
method Framework using guidance functions for discrete data.
result Demonstrated utility on molecular graph generation tasks.
Paper proposes a probabilistic alignment method for domain adaptation.
problem Latent distribution mismatch and miscalibrated uncertainty in adapting large-scale models.
method Bayesian latent transport framework with PAC-Bayesian regularization.
result Reduction in latent manifold discrepancy and improved uncertainty calibration.
Hybrid deep architectures with reasoning layers show promising convergence and generalization properties.
problem Understanding the theoretical foundations of hybrid deep architectures with reasoning layers.
method Analyzing the interplay between algorithm layers and neural components in deep architectures.
result Properties of algorithm layers are closely related to the approximation and generalization abilities of end-to-end models.
This study evaluates different normalizing flow architectures for MCMC.
problem Lack of systematic comparison of normalizing flow architectures in MCMC.
method Extensive evaluation of various normalizing flow architectures on different MCMC methods and target distributions.
result Contractive residual flows are the best general-purpose models for MCMC.
TempoPFN models for zero-shot time series forecasting using synthetic data.
problem Efficient long-horizon prediction and reproducibility in zero-shot time series forecasting.
method Linear RNNs pre-trained on synthetic data with GatedDeltaProduct architecture and state-weaving.
result Achieves top-tier competitive performance on various benchmarks.
This paper organizes recent deep learning theory advances.
problem Lack of theoretical foundations in deep learning.
method Literature review and categorization into six groups.
result Organized recent advances in deep learning theory.
DAG-FM discovers causal relationships from heterogeneous data.
problem Challenges in causal discovery from heterogeneous causal mechanisms.
method DAG-FM uses two specialized Transformer-based sub-modules and a robust tabular interaction block to model complex row-column interactions.
result DAG-FM achieves state-of-the-art performance on synthetic and real-world datasets.
PRAGMA models financial event sequences for various banking tasks.
problem Handling diverse financial data for multiple applications.
method Pre-training a Transformer model on a large banking event corpus with a self-supervised objective.
result PRAGMA achieves superior performance across multiple financial domains from raw event sequences.
We provide a characterization of two types of directed homology for fully-connected, feedforward neural network architectures. These exact characterizations of the directed homology structure of a neural network architecture are the first of their kind. We show that the directed flag homology of deep networks reduces t…
Study simulates liquidity in fractional ownership markets using ABM.
problem Understanding liquidity dynamics in illiquid markets.
method Agent-based modeling (ABM) with empirical data.
result Simulation reveals insights into market structures and trading behaviors.
Measures equivariance in vision models using Lie derivative.
problem Understanding the role of equivariance in recent vision models.
method Introducing Lie derivative to measure equivariance with strong mathematical foundations and minimal hyperparameters.
result Many violations of equivariance can be linked to spatial aliasing in network layers, and larger models tend to display more equivariance.
Survey of AI in finance covering models, strategies, and knowledge systems.
problem Challenges in applying AI to financial markets, especially in high-frequency trading.
method Systematic analysis of financial AI across predictive models, decision frameworks, and knowledge augmentation systems.
result Critical trade-offs and gaps between theoretical advances and practical implementation in financial AI.
Stabilizing black-box algorithms through task-oriented randomization
problem Ensuring stability of black-box models
method Task-oriented randomization
result Established rigorous theoretical foundations and demonstrated effectiveness through simulations and real-world applications
Paper defines XAI concepts using category theory.
problem Lack of precise mathematical definitions for XAI.
method Uses Category theory to define XAI concepts rigorously.
result Establishes a theoretical foundation for XAI.
uGMM-NN integrates probabilistic reasoning into neural networks.
problem Capturing multimodality and uncertainty in neural network activations.
method Parameterizes activations as univariate Gaussian mixtures with learnable parameters.
result Competitive discriminative performance with probabilistic activations.
A model learns causal graphs from summary statistics of synthetic data.
problem Causal discovery algorithms are brittle with large sets of variables and limited data.
method A supervised model trained on synthetic data predicts causal graphs from summary statistics.
result The model generalizes well beyond its training set and runs on large graphs.
This work benchmarks and theorizes robust NAS under adversarial training.
problem Lack of benchmark evaluations and theoretical guarantees for robust NAS architectures under adversarial training.
method Released a comprehensive data set and established a generalization theory using the neural tangent kernel.
result Established a generalization theory for robust NAS architectures under adversarial training.
Residual neural networks don't help overcome sampling complexity issues.
problem Learning invertible residual neural networks from samples is hard due to the curse of dimensionality.
method Investigated invertible residual neural networks and their sampling complexity.
result Invertible residual neural networks still suffer from the curse of dimensionality in sampling complexity.
Kronos improves financial time series analysis with a pre-trained model.
problem Limited application of large-scale models to financial candlestick data.
method Unified, scalable pre-training framework for financial K-line modeling.
result Kronos excels in financial tasks like price forecasting and volatility prediction.
We present an approach for reconfiguration of dynamic visual sensor networks with deep reinforcement learning (RL). Our RL agent uses a modified asynchronous advantage actor-critic framework and the recently proposed Relational Network module at the foundation of its network architecture. To address the issue of sample…
The paper explores neural scaling laws for deep operator networks, offering a theoretical foundation.
problem Understanding neural scaling laws in deep operator networks.
method Theoretical analysis of approximation and generalization errors.
result Established a theoretical framework to quantify neural scaling laws for deep operator networks.
Researchers analyze neural process architectures and their representational capacities.
problem Understanding what functions can be represented by different neural process architectures.
method Analyzing four types of neural process architectures: CNPs, ANPs, TNPs, and their latent variants.
result Prove these architectures form a strict hierarchy and characterize their representational capabilities.
A new optimisation framework for neural networks without hyperparameters.
problem Lack of explicit architectural information in optimisation frameworks.
method Extends mirror descent to account for neural architecture, transforming a Bregman divergence.
result Automatic gradient descent: a first-order optimiser without hyperparameters.
Neural SVEs model complex systems with memory, outperforming traditional methods.
problem Modeling systems with memory effects and irregular behavior.
method Introducing neural stochastic Volterra equations as a physics-inspired architecture.
result Neural SVEs outperform neural SDEs and DeepONets in various applications.
Revises Bayesian model averaging for foundation models.
problem Ensemble pre-trained and lightly-finetuned foundation models for improved classification performance.
method Introduces trainable linear classifiers and computationally cheaper model averaging scheme (OMA).
result Ensembled models can better predict on various datasets.
TSFMs improve financial forecasting across diverse tasks with strong transferability.
problem Complex nonlinear relationships, temporal dependencies, and limited data in financial time series forecasting.
method Pretraining on diverse time series corpora followed by task-specific adaptation.
result Tiny Time Mixers (TTM) achieved 25-50% better performance on limited data and 15-30% improvements on longer datasets.
CDFM aims to unify causal discovery across diverse datasets.
problem Fragmented, test-driven causal discovery approaches struggle with modern data heterogeneity.
method CDFM is a unified, general-purpose framework using a variational decomposition of causal mechanisms.
result CDFM outperforms traditional algorithms across diverse datasets.
Time series foundation models are well-calibrated, improving over baseline models.
problem Calibration of time series foundation models for practical applications.
method Systematic evaluations of five time series foundation models and two baselines, assessing calibration, prediction heads, and long-term forecasting.
result Time series foundation models are consistently better calibrated than baseline models and do not show over- or under-confidence.
Introduces foundation priors for using model-generated data in empirical research.
problem Using model-generated data as real observations in empirical research.
method Introduces foundation priors as an exponential-tilted, generalized Bayesian update of the user's primitive prior.
result Synthetic data reflects both model patterns and user's priors, enabling principled use in empirical work.
A new method uses algebraic insights to create approximately equivariant networks without complex architectures.
problem Designing equivariant neural networks with complex architectures and high computational cost.
method Imposes the group's regular representation as an inductive bias via an auxiliary loss, adding no learnable parameters.
result Matches or outperforms specialized models in several cases, even for infinite groups.