Searcharxiv⌕ Search

arXiv subjects

Dmitry Efimov

Publications and source records attributed to Dmitry Efimov.

17 recordsLinked to original sources
arXiv2026

Bootstrap-Conditioned Action Selection with Tabular Foundation Models

Contextual bandits offer a natural framework for sample-efficient personalization, but practical deployment remains difficult under sparse, biased interaction data, unreliable uncertainty estimates, and severe cold starts. We study whether pre-trained tabular foundation models with in-context learning can be turned into randomized policies for online decision making. We propose BC-ICL (Bootstrap-conditioned action selection using ICL), which at each round draws a bootstrap resample of the interaction history, conditions a frozen pre-trained ICL model on that resample, scores all actions, and selects the action with the highest sampled score. We further introduce an arm-context conditioning architecture that promotes shared statistical strength across actions and helps avoid common bootstrap failure modes of isolated-arm bandits. Empirically, this policy delivers strong early-round regret and regret performance on standard contextual bandit suites, outperforming established baselines under a strict online protocol.

cs.LG↗
arXiv2026

Time-Aware Prior Fitted Networks for Zero-Shot Forecasting with Exogenous Variables

In many time series forecasting settings, the target time series is accompanied by exogenous covariates, such as promotions and prices in retail demand; temperature in energy load; calendar and holiday indicators for traffic or sales; and grid load or fuel costs in electricity pricing. Ignoring these exogenous signals can substantially degrade forecasting accuracy, particularly when they drive spikes, discontinuities, or regime and phase changes in the target series. Most current time series foundation models (e.g., Chronos, Sundial, TimesFM, TimeMoE, TimeLLM, and LagLlama) ignore exogenous covariates and make forecasts solely from the numerical time series history, thereby limiting their performance. In this paper, we develop ApolloPFN, a prior-data fitted network (PFN) that is time-aware (unlike prior PFNs) and that natively incorporates exogenous covariates (unlike prior univariate forecasters). Our design introduces two major advances: (i) a synthetic data generation framework that injects realistic temporal patterns, structural changes, and exogenous dependencies into the PFN prior; and (ii) time-aware architectural modifications that embed inductive biases needed to exploit temporal context. We demonstrate that ApolloPFN outperforms existing baselines across several forecasting benchmarks with exogenous information, including M5, electric price forecasting, UCI Air Quality, and Solar Energy datasets.

cs.LG↗
arXiv2026

Zero-shot Forecasting by Simulation Alone

Zero-shot time-series forecasting holds great promise, but is still in its infancy, hindered by limited and biased data corpora, leakage-prone evaluation, and privacy and licensing constraints. Motivated by these challenges, we propose the first practical univariate time series simulation pipeline which is simultaneously fast enough for on-the-fly data generation and enables notable zero-shot forecasting performance on M-Series and GiftEval benchmarks that capture trend/seasonality/intermittency patterns, typical of industrial forecasting applications across a variety of domains. Our simulator, which we call SarSim0 (SARIMA Simulator for Zero-Shot Forecasting), is based off of a seasonal autoregressive integrated moving average (SARIMA) model as its core data source. Due to instability in the autoregressive component, naive SARIMA simulation often leads to unusable paths. Instead, we follow a three-step procedure: (1) we sample well-behaved trajectories from its characteristic polynomial stability region; (2) we introduce a superposition scheme that combines multiple paths into rich multi-seasonality traces; and (3) we add rate-based heavy-tailed noise models to capture burstiness and intermittency alongside seasonalities and trends. SarSim0 is orders of magnitude faster than kernel-based generators, and it enables training on circa 1B unique purely simulated series, generated on the fly; after which well-established neural network backbones exhibit strong zero-shot generalization, surpassing strong statistical forecasters and recent foundation baselines, while operating under strict zero-shot protocol. Notably, on GiftEval we observe a "student-beats-teacher" effect: models trained on our simulations exceed the forecasting accuracy of the AutoARIMA generating processes.

cs.LG↗
arXiv2025

Forking-Sequences: Statistically and Computationally Efficient Multi-Horizon Forecasting with Reduced Volatility

While accuracy is a critical requirement for time series forecasting, an equally important desideratum is reasonable forecast volatility across forecast creation dates (FCDs). Even highly accurate models can produce erratic revisions between FCDs, undermining trust and disrupting downstream decision-making. To improve the volatility of forecast revisions, state-of-the-art models like MQCNN, MQT, and SPADE employ a powerful yet underexplored neural network architectural design: forking-sequences. This architectural design jointly encodes and decodes the entire time series across all FCDs, producing an entire multi-horizon forecast grid in a single forward pass. This approach contrasts with conventional neural forecasting methods that process FCDs independently, generating only a single multi-horizon forecast per forward pass. In this work, we formalize the forking-sequences design and motivate its broader adoption by introducing a metric for quantifying excess volatility in forecast revisions and by providing theoretical and empirical analysis. We theoretically motivate three key benefits of forking-sequences: (i) reduced forecast volatility through ensembling; (ii) gradient variance reduction, improving the statistical efficiency of the training procedure; and (iii) improved inference computational efficiency. We validate the benefits of forking-sequences compared to baseline window-sampling on the M-series benchmark, using 16 datasets from the M1, M3, M4, and Tourism competitions. We observe median sCRPS improvements across datasets of 46.2%, 49.3%, 28.6%, 24.7%, and 6.4% for RNN, LSTM, CNN, Transformer, and State Space-based architectures, respectively. We then show that forecast ensembling during inference can reduce median forecast volatility by 13.2%, 13.0%, 10.9%, 10.2%, and 11.2% for these respective models trained with forking-sequences, while maintaining accuracy.

cs.LG↗
arXiv2025

A More Realistic Evaluation of Cross-Frequency Transfer Learning and Foundation Forecasting Models

Cross-frequency transfer learning (CFTL) has emerged as a popular framework for curating large-scale time series datasets to pre-train foundation forecasting models (FFMs). Although CFTL has shown promise, current benchmarking practices fall short of accurately assessing its performance. This shortcoming stems from many factors: an over-reliance on small-scale evaluation datasets; inadequate treatment of sample size when computing summary statistics; reporting of suboptimal statistical models; and failing to account for non-negligible risks of overlap between pre-training and test datasets. To address these limitations, we introduce a unified reimplementation of widely-adopted neural forecasting networks, adapting them for the CFTL setup; we pre-train only on proprietary and synthetic data, being careful to prevent test leakage; and we evaluate on 15 large, diverse public forecast competition datasets. Our empirical analysis reveals that statistical models' accuracy is frequently underreported. Notably, we confirm that statistical models and their ensembles consistently outperform existing FFMs by more than 8.2% in sCRPS, and by more than 20% MASE, across datasets. However, we also find that synthetic dataset pre-training does improve the accuracy of a FFM by 7% percent.

cs.LG↗
arXiv2025

Probabilistic Pretraining for Neural Regression

Transfer learning for probabilistic regression remains underexplored. This work closes this gap by introducing NIAQUE, Neural Interpretable Any-Quantile Estimation, a new model designed for transfer learning in probabilistic regression through permutation invariance. We demonstrate that pre-training NIAQUE directly on diverse downstream regression datasets and fine-tuning it on a specific target dataset enhances performance on individual regression tasks, showcasing the positive impact of probabilistic transfer learning. Furthermore, we highlight the effectiveness of NIAQUE in Kaggle competitions against strong baselines involving tree-based models and recent neural foundation models TabPFN and TabDPT. The findings highlight NIAQUE's efficacy as a robust and scalable framework for probabilistic regression, leveraging transfer learning to enhance predictive performance.

cs.LG↗
arXiv2025

TAT: Temporal-Aligned Transformer for Multi-Horizon Peak Demand Forecasting

Multi-horizon time series forecasting has many practical applications such as demand forecasting. Accurate demand prediction is critical to help make buying and inventory decisions for supply chain management of e-commerce and physical retailers, and such predictions are typically required for future horizons extending tens of weeks. This is especially challenging during high-stake sales events when demand peaks are particularly difficult to predict accurately. However, these events are important not only for managing supply chain operations but also for ensuring a seamless shopping experience for customers. To address this challenge, we propose Temporal-Aligned Transformer (TAT), a multi-horizon forecaster leveraging apriori-known context variables such as holiday and promotion events information for improving predictive performance. Our model consists of an encoder and decoder, both embedded with a novel Temporal Alignment Attention (TAA), designed to learn context-dependent alignment for peak demand forecasting. We conduct extensive empirical analysis on two large-scale proprietary datasets from a large e-commerce retailer. We demonstrate that TAT brings up to 30% accuracy improvement on peak demand forecasting while maintaining competitive overall performance compared to other state-of-the-art methods.

cs.LG↗
arXiv2025

On efficiency of double ionization of a three-electron atom in moderate laser field intensities

We study double ionization in a one-dimensional model of a three-electron atom exposed to strong short laser fields. The corresponding ab initio simulations reveal a characteristic ``knee'' in the ionization yield at lower field intensities than in comparable two-electron systems. To understand the origin of this shift, a quantitative semi-analytic model for double ionization of a three-electron atom has been created. In this model, ionization is treated as a combination of different scenarios and channels. Each channel is simulated with either Quantitative Rescattering Theory (QRS) or two-electron ab initio model. Analysis of the semi-analytical results indicates that the shifted part of the ``knee'' is dominated by a recollision-excitation channel in which the rescattering electron initially escapes the parent atom from an inner orbital. This counterintuitive phenomenon is then discussed and explained.

physics.atom-ph↗
arXiv2024

LLMForecaster: Improving Seasonal Event Forecasts with Unstructured Textual Data

Modern time-series forecasting models often fail to make full use of rich unstructured information about the time series themselves. This lack of proper conditioning can lead to obvious model failures; for example, models may be unaware of the details of a particular product, and hence fail to anticipate seasonal surges in customer demand in the lead up to major exogenous events like holidays for clearly relevant products. To address this shortcoming, this paper introduces a novel forecast post-processor -- which we call LLMForecaster -- that fine-tunes large language models (LLMs) to incorporate unstructured semantic and contextual information and historical data to improve the forecasts from an existing demand forecasting pipeline. In an industry-scale retail application, we demonstrate that our technique yields statistically significantly forecast improvements across several sets of products subject to holiday-driven demand surges.

cs.LG↗
arXiv2024

$\spadesuit$ SPADE $\spadesuit$ Split Peak Attention DEcomposition

Demand forecasting faces challenges induced by Peak Events (PEs) corresponding to special periods such as promotions and holidays. Peak events create significant spikes in demand followed by demand ramp down periods. Neural networks like MQCNN and MQT overreact to demand peaks by carrying over the elevated PE demand into subsequent Post-Peak-Event (PPE) periods, resulting in significantly over-biased forecasts. To tackle this challenge, we introduce a neural forecasting model called Split Peak Attention DEcomposition, SPADE. This model reduces the impact of PEs on subsequent forecasts by modeling forecasting as consisting of two separate tasks: one for PEs; and the other for the rest. Its architecture then uses masked convolution filters and a specialized Peak Attention module. We show SPADE's performance on a worldwide retail dataset with hundreds of millions of products. Our results reveal an overall PPE improvement of 4.5%, a 30% improvement for most affected forecasts after promotions and holidays, and an improvement in PE accuracy by 3.9%, relative to current production models.

cs.LG↗
arXiv2021

Three-electron correlations in strong laser field ionization: Spin induced effects

Strong field processes in the non-relativistic regime are insensitive to the electron spin, i.e. the observables appear to be independent of this electron property. This does not have to be the case for several active electrons where Pauli principle may affect the their dynamics. We exemplify this statement studying model atoms with three active electrons interacting with strong pulsed radiation, using an ab-initio time-dependent Schrödinger equation on a grid. In our restricted dimensionality model we are able, for the first time, to analyse momenta correlations of the three outgoing electrons using Dalitz plots. We show that significant differences are obtained between model Neon and Nitrogen atoms. These differences are traced back to the different symmetries of the electronic wavefunctions, and directly related to the different initial state spin components.

physics.atom-ph↗
arXiv2021

Hafnian of two-parameter matrices

The concept of the hafnian first appeared in the works on quantum field theory by E. R. Caianiello. However, it also has an important combinatorial property: the hafnian of the adjacency matrix of an undirected weighted graph is equal to the total sum of the weights of perfect matchings in this graph. In general, the use of the hafnian is limited by the complexity of its computation. In this paper, we present an efficient method for the exact calculation of the hafnian of two-parameter matrices. In terms of graphs, we count the total sum of the weights of perfect matchings in graphs whose edge weights take only two values. This method is based on the formula expressing the hafnian of a sum of two matrices through the product of the hafnians of their submatrices. The necessary condition for the application of this method is the possibility to count the number of k-edge matchings in some graphs. We consider two special cases in detail using a Toeplitz matrix as the two-parameter matrix. As an example, we propose a new interpretation of some of the sequences from the On-Line Encyclopedia of Integer Sequences and then provide new analytical formulas to count the number of certain linear chord diagrams.

math.CO↗
arXiv2020

Sequential Deep Learning for Credit Risk Monitoring with Tabular Financial Data

Machine learning plays an essential role in preventing financial losses in the banking industry. Perhaps the most pertinent prediction task that can result in billions of dollars in losses each year is the assessment of credit risk (i.e., the risk of default on debt). Today, much of the gains from machine learning to predict credit risk are driven by gradient boosted decision tree models. However, these gains begin to plateau without the addition of expensive new data sources or highly engineered features. In this paper, we present our attempts to create a novel approach to assessing credit risk using deep learning that does not rely on new model inputs. We propose a new credit card transaction sampling technique to use with deep recurrent and causal convolution-based neural networks that exploits long historical sequences of financial data without costly resource requirements. We show that our sequential deep learning approach using a temporal convolutional network outperformed the benchmark non-sequential tree-based model, achieving significant financial savings and earlier detection of credit risk. We also demonstrate the potential for our approach to be used in a production environment, where our sampling technique allows for sequences to be stored efficiently in memory and used for fast online learning and inference.

q-fin.RM↗
arXiv2020

Using generative adversarial networks to synthesize artificial financial datasets

Generative Adversarial Networks (GANs) became very popular for generation of realistically looking images. In this paper, we propose to use GANs to synthesize artificial financial data for research and benchmarking purposes. We test this approach on three American Express datasets, and show that properly trained GANs can replicate these datasets with high fidelity. For our experiments, we define a novel type of GAN, and suggest methods for data preprocessing that allow good training and testing performance of GANs. We also discuss methods for evaluating the quality of generated data, and their comparison with the original real data.

cs.LG↗
arXiv2017

Determinants of generalized binary band matrices

Under binary matrices we mean matrices whose entries take one of two values. In this paper, explicit formulae for calculating the determinant of some type of binary Toeplitz matrices are obtained. Examples of the application of the determinant of binary Toeplitz matrices for the enumeration of even and odd permutations of different types are given.

math.RA↗
arXiv2009

On exact representations of the motions group of Galilean plane

The Pimenov algebra with two generators is defined and some of its properties are shown. Some exact matrix over the Pimenov algebra representations of the motions group of Galilean plane (the Galilean group) are considered. A geometric interpretation of them is giving. We consider also a exact representation of the Galilean group by elements of Grassmann algebra.

math.RT↗