A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
Specialists outperform generalists in ensemble classification.
Study identifies specialist representations from generalist models without parametric constraints.
Generalist neural learner can execute multiple algorithms.
Several recent deep neural networks experiments leverage the generalist-specialist paradigm for classification. However, no formal study compared the performance of different clustering algorithms for class assignment. In this paper we perform such a study, suggest slight modifications to the clustering procedures, and…
ContextFlow++ improves generative models by conditioning on mixed-variable contexts.
Develops a framework for multi-objective learning in diffusion models with limited labeled data.
Specialists tolerate defects to gain flexibility, which can be removed when needed.
Specialists tolerate defects to gain flexibility, which can be removed when needed.
This is an invited article for the Discussion and Debate special issue of The European Physical Journal Special Topics on the subject "Can Economics Be a Physical Science?" The first part of the paper traces the personal path of the author from theoretical physics to economics. It briefly summarizes applications of sta…
Framework improves ETF volatility forecasting by adapting to market conditions.
This paper considers the problem of choosing a good classifier. For each problem there exist an optimal classifier, but none are optimal, regarding the error rate, in all cases. Because there exists a large number of classifiers, a user would rather prefer an all-purpose classifier that is easy to adjust, in the hope t…
mvlearn simplifies multiview machine learning for non-specialists.
Study on academic finance evolution over 30 years.
We address the problem of predicting the labeling of a graph in an online setting when the labeling is changing over time. We present an algorithm based on a specialist approach; we develop the machinery of cluster specialists which probabilistically exploits the cluster structure in the graph. Our algorithm has two va…
MAP-Elites generates diverse trading strategies for improved execution performance.
Shared workspace improves neural module coordination in deep learning.
RG-TTA adapts neural forecasters to streaming time series shifts by modulating adaptation intensity.
LLM Pro Finance Suite enhances financial NLP with instruction-tuned models.
We provide an (almost) self-contained construction of the Witten-Reshetikhin-Turaev representations of the mapping class group. We describe its properties including its Hermitian structure, irreducibility and integrality (at prime level). The construction of these notes relies only on skein theory (Kauffman Bracket) an…
VR game data for P300 BCI with raccoon vs demon stimuli.
We present a model that investigates the spontaneous emergence of randomness in equity market microstructure. The phase space analysis of our model exposes an endogenous source of fluctuation in price and volume. We formulate a control problem for maximizing price regularity and stability while minimizing entanglement …
Electroencephalography (EEG) is an extensively-used and well-studied technique in the field of medical diagnostics and treatment for brain disorders, including epilepsy, migraines, and tumors. The analysis and interpretation of EEGs require physicians to have specialized training, which is not common even among most do…
We define a non-absolutely convergent integration on integral currents of dimension 1 in Euclidean space. This integral is closely related to the Henstock-Kurzweil and Pfeffer Integrals. Using it, we prove a generalized Fundamental Theorem of Calculus on these currents. A detailed presentation of Henstock-Kurzweil Inte…
A very simple way to improve the performance of almost any machine learning algorithm is to train many different models on the same data and then to average their predictions. Unfortunately, making predictions using a whole ensemble of models is cumbersome and may be too computationally expensive to allow deployment to…
With the advent of smartphone indirect ophthalmoscopy, teleophthalmology - the use of specialist ophthalmology assets at a distance from the patient - has experienced a breakthrough, promising enormous benefits especially for healthcare in distant, inaccessible or opthalmologically underserved areas, where specialists …
Proposes a new framework for uncertainty-aware LLM post-training.
Confidence-based deferral works well in many scenarios but fails in specific cases.
Automatic summarization of natural language is a current topic in computer science research and industry, studied for decades because of its usefulness across multiple domains. For example, summarization is necessary to create reviews such as this one. Research and applications have achieved some success in extractive …
New method for evaluating LLMs reduces bias in open-ended evaluations.
We considered observational data available from the MIMIC-III open-access ICU database and collected within a study period between year 2002 up to 2011. If a patient had multiple admissions to the ICU during the 30 days before death, only the first stay was analyzed, leading to a final set of 6,436 unique ICU admission…
Achilles predicts Gold vs USD with a profitable trading bot.
Kernel interpolation improved with continuous volume sampling.
VQShape learns interpretable time-series representations and achieves comparable performance to specialist models.
MRC improves credit assignment in multi-agent LLM systems, achieving high returns and transparency.
Global and local blowups of manifolds are proven equivalent.
Recently, classical results on completeness of trajectories of Hamiltonian systems obtained at the beginning of the seventies, have been revisited, improved and applied to Lorentzian Geometry. Our aim here is threefold: to give explicit proofs of some technicalities in the background of the specialists, to show that th…
Foundation models outperform supervised methods in time series forecasting across various operational regimes.
Interactive art piece shows braid groups and plane motions.
We price weather-contingent options by use of Monte Carlo simulations. After calibrating the models to fit quoted prices, we analyze bid-ask spreads in terms of correlations across markets. Results are presented for a double-trigger Weather vs. Natural Gas call option.
Our work focuses on the problem of predicting the transfer of pediatric patients from the general ward of a hospital to the pediatric intensive care unit. Using data collected over 5.5 years from the electronic health records of two medical facilities, we develop classifiers based on adaptive boosting and gradient tree…
Bayesian network structures are usually built using only the data and starting from an empty network or from a naive Bayes structure. Very often, in some domains, like medicine, a prior structure knowledge is already known. This structure can be automatically or manually refined in search for better performance models.…
Survey simplifies embedding theorems for manifolds.
Proposes SOVR loss to improve adversarial robustness by increasing logit margins.
We present Vector-Space Markov Random Fields (VS-MRFs), a novel class of undirected graphical models where each variable can belong to an arbitrary vector space. VS-MRFs generalize a recent line of work on scalar-valued, uni-parameter exponential family and mixed graphical models, thereby greatly broadening the class o…
Training of one-vs.-rest SVMs can be parallelized over the number of classes in a straight forward way. Given enough computational resources, one-vs.-rest SVMs can thus be trained on data involving a large number of classes. The same cannot be stated, however, for the so-called all-in-one SVMs, which require solving a …
We present a framework for compactly summarizing many recent results in efficient and/or biologically plausible online training of recurrent neural networks (RNN). The framework organizes algorithms according to several criteria: (a) past vs. future facing, (b) tensor structure, (c) stochastic vs. deterministic, and (d…
Many supervised learning tasks are emerged in dual forms, e.g., English-to-French translation vs. French-to-English translation, speech recognition vs. text to speech, and image classification vs. image generation. Two dual tasks have intrinsic connections with each other due to the probabilistic correlation between th…
The paper revisits discriminative vs. generative classifiers, showing naive Bayes requires fewer samples.