arXiv research

This paper compares two loss functions for learning from aggregated responses and introduces an interpolating estimator.

The paper addresses privacy in rank aggregation using randomized responses.

Finding the most effective way to aggregate multi-subject fMRI data is a long-standing and challenging problem. It is of increasing interest in contemporary fMRI studies of human cognition due to the scarcity of data per subject and the variability of brain anatomy and functional response across subjects. Recent work o…

The paper introduces BCART models for aggregate claim amount, improving frequency-severity and joint modeling.

Paper analyzes sparse aggregation in GLMs with Kullback-Leibler risk bounds.

Study robust linear regression without distributional assumptions for heavy-tailed responses.

The literature of heavy tails (typically) starts with a random walk and finds mechanisms that lead to fat tails under aggregation. We follow the inverse route and show how starting with fat tails we get to thin-tails when deriving the probability distribution of the response to a random variable. We introduce a general…

Study reveals LLM personas have two distinct components: frame-robust aggregated traits and frame-dependent geometric features.

The paper proposes a new method to evaluate LLM agent responses using ECDF clustering.

Unified analytic account of correlation emergence and Epps effect in coupled limit order books

Study examines how industrial emissions evolve over time in response to various factors.

Label aggregation makes learning robust to noisy labels.

Method fuses low and high-resolution data for better health estimates.

Paper optimizes demand aggregation for low-level electricity markets.

Unified framework for binary responses using AUC loss and low-rank constraint.

Calibrated models can lead to miscalibrated aggregations in strategic interactions.

Noisy labeled data is more a norm than a rarity for crowd sourced contents. It is effective to distill noise and infer correct labels through aggregation results from crowd workers. To ensure the time relevance and overcome slow responses of workers, online label aggregation is increasingly requested, calling for solut…

MELO predicts electricity loads by adapting to shifts without external indicators.

CoT-UQ improves LLM uncertainty quantification by integrating reasoning steps.

UTOPIA aggregates multiple prediction intervals efficiently.

We present a class of macroscopic models of the Limit Order Book to simulate the aggregate behaviour of market makers in response to trading flows. The resulting models are solved numerically and asymptotically, and a class of similarity solutions linked to order book formation and recovery is explored. The main result…

New method decomposes local projections to reveal historical drivers of estimates.

Price responsiveness is a major feature of end use customers (EUCs) that participate in demand response (DR) programs, and has been conventionally modeled with static demand functions, which take the electricity price as the input and the aggregate energy consumption as the output. This, however, neglects the inherent …

A new method aggregates generative classifiers to resist adversarial attacks.

We introduce a new tool for interpreting neural net responses, namely full-gradients, which decomposes the neural net response into input sensitivity and per-neuron sensitivity components. This is the first proposed representation which satisfies two key properties: completeness and weak dependence, which provably cann…

Response time improves alignment with diverse human preferences.

LLMs can predict CFO responses to economic surveys

New methods protect privacy while providing accurate prediction sets.

Online crowdsourcing provides a scalable and inexpensive means to collect knowledge (e.g. labels) about various types of data items (e.g. text, audio, video). However, it is also known to result in large variance in the quality of recorded responses which often cannot be directly used for training machine learning syst…

Localized Multidirectional Correction improves non-refusal target-response behavior in foundation models.

VED framework learns low-dimensional latent representations of physical systems.

New framework handles dynamic contexts in reinforcement learning.

TransNet improves community detection on target networks using privacy-preserved source networks.

Paper addresses privacy and robustness in stochastic linear bandits.

The dual crises of the sub-prime mortgage crisis and the global financial crisis has prompted a call for explanations of non-equilibrium market dynamics. Recently a promising approach has been the use of agent based models (ABMs) to simulate aggregate market dynamics. A key aspect of these models is the endogenous emer…

E2Tree explains random forest models in regression tasks.

Model trains passing events on a bridge using multilevel Gaussian process.

AAggFF improves federated learning fairness through sequential decision making.

Behavior of systems that are functions of anticipated behavior of other systems, whose own behavior is also anticipatory but homeostatic and determined by hierarchical ordering, which changes over time, of sets of possible environments that are not co-possible, is proven to be highly non-linear and sensitively dependen…

Generative Adversarial Networks simulate realistic market interactions.

Framework controls uncertainty in LLMs without labels or probabilities.

Determining whether hypotensive patients in intensive care units (ICUs) should receive fluid bolus therapy (FBT) has been an extremely challenging task for intensive care physicians as the corresponding increase in blood pressure has been hard to predict. Our study utilized regression models and attention-based recurre…

PCL framework optimizes climate risk management across three clusters.

Study optimal policy regret in partially observable Markov games with adaptive opponents.

We introduce a framework to study the effective objectives at different time scales of financial market microstructure. The financial market can be regarded as a complex adaptive system, where purposeful agents collectively and simultaneously create and perceive their environment as they interact with it. It has been s…

A new ICA model identifies shared brain activity patterns across subjects.

Machine learning research for developing countries can demonstrate clear sustainable impact by delivering actionable and timely information to in-country government organisations (GOs) and NGOs in response to their critical information requirements. We co-create products with UK and in-country commercial, GO and NGO pa…

When training clinical prediction models from electronic health records (EHRs), a key concern should be a model's ability to sustain performance over time when deployed, even as care practices, database systems, and population demographics evolve. Due to de-identification requirements, however, current experimental pra…