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.
Scores measure certainty and doubt in classification predictions.
We take a look the changes of different asset prices over variable periods, using both traditional and spectral methods, and discover universality phenomena which hold (in some cases) across asset classes.
Recent news cast doubts on London Interbank Offered Rate (LIBOR) integrity. Given its economic importance and the delay with which authorities realize about this situation, we aim to find an objective method in order to detect departures in the LIBOR rate that from the expected behavior. We analyze several interest rat…
Using very weak criteria for what may constitute a noncommutative geometry, I show that a pseudo-Riemannian manifold can only be smoothly deformed into noncommutative geometries if certain geometric obstructions vanish. These obstructions can be expressed as a system of partial differential equations relating the metri…
In this note we revisit and extend few classical and recent results on the definition and use of the Futaki invariant in connection with the existence problem for Kaehler constant scalar curvature metrics on polarized algebraic manifolds, especially in the case of resolution of singularities. The general inspiration be…
Workshop notes on positivity in Lie groups and its applications.
According to the definition of the London Interbank Offered Rate (LIBOR), contributing banks should give fair estimates of their own borrowing costs in the interbank market. Between 2007 and 2009, several banks made inappropriate submissions of LIBOR, sometimes motivated by profit-seeking from their trading positions. …
Unified HS and related methods with explicit modeling assumptions.
Two classes of gamma-ray bursts (GRBs), short and long, have been determined without any doubts, and are usually ascribed to different progenitors, yet these classes overlap for a variety of descriptive parameters. A subsample of 46 long and 22 short GRBs with estimated Hurst Exponents (HEs), complemented by mi…
Weight decay is one of the standard tricks in the neural network toolbox, but the reasons for its regularization effect are poorly understood, and recent results have cast doubt on the traditional interpretation in terms of regularization. Literal weight decay has been shown to outperform regularization for…
Diffusion models achieve nearly optimal distribution estimation in various spaces.
Unbiased methods for alpha-divergence minimization struggle in high dimensions.
The biological plausibility of the backpropagation algorithm has long been doubted by neuroscientists. Two major reasons are that neurons would need to send two different types of signal in the forward and backward phases, and that pairs of neurons would need to communicate through symmetric bidirectional connections. …
This paper is a sequel of "Solvable symmetric black hole in anti de Sitter spaces" [arXiv:math.DG/0510442]. In the latter, we described the BTZ black hole in every dimension by defining the singularity as the closed orbits of the Iwasawa subgroup of SO(2,n). In this article, we study the horizon of the black hole and w…
Paper tackles selective regression using uncertainty estimation.
Recent years have witnessed widespread adoption of machine learning (ML)/deep learning (DL) techniques due to their superior performance for a variety of healthcare applications ranging from the prediction of cardiac arrest from one-dimensional heart signals to computer-aided diagnosis (CADx) using multi-dimensional me…
Adaptive Group Lasso selects important features in neural networks.
The repeated community-wide reuse of test sets in popular benchmark problems raises doubts about the credibility of reported test-error rates. Verifying whether a learned model is overfitted to a test set is challenging as independent test sets drawn from the same data distribution are usually unavailable, while other …
New method explains predictive uncertainty by focusing on second-order effects.
The first part of this article intends to present the role played by Thom in diffusing Smale's ideas about immersion theory, at a time (1957) where some famous mathematicians were doubtful about them: it is clearly impossible to make the sphere inside out! Around a decade later, M. Gromov transformed Smale's idea in wh…
New findings suggest non-contrastive learning has many bad minima, not just collapsed ones.
Three new types of graded Lie groups are constructed and analyzed.
Study reveals limitations of fair representation learning methods and cautions against their use in performance-sensitive tasks.
This paper suggests claim history will be deprecated in future auto insurance rates.
Paper proposes a new Autoencoder for robustly encoding white matter streamlines.
e-Yantra Robotics Competition (eYRC) is a unique Robotics Competition hosted by IIT Bombay that is actually an Embedded Systems and Robotics MOOC. Registrations have been growing exponentially in each year from 4500 in 2012 to over 34000 in 2019. In this 5-month long competition students learn complex skills under seve…
This work identifies and mitigates reasoning shortcuts in Neuro-Symbolic models.
Deep ResNets exhibit distinct scaling properties with depth, challenging neural ODE models.
Study compares adversarial regularization to sole supervision in machine learning.
Poisoning datasets can reveal private details of other users' training points.
Interpreting predictions from tree ensemble methods such as gradient boosting machines and random forests is important, yet feature attribution for trees is often heuristic and not individualized for each prediction. Here we show that popular feature attribution methods are inconsistent, meaning they can lower a featur…
New risk measure uses Fourier analysis of stock prices.
Study validates ML-UQ calibration statistics using simulated reference values.
New tuning rules for Metropolis algorithms derived from Bayesian large-sample asymptotics.
New CNN initialization scheme derived from modern architectures.
New method proves neural networks can select features consistently.
Aimed at explaining the surprisingly good generalization behavior of overparameterized deep networks, recent works have developed a variety of generalization bounds for deep learning, all based on the fundamental learning-theoretic technique of uniform convergence. While it is well-known that many of these existing bou…
New loss improves DNN calibration without sacrificing accuracy.
Recently, the efficiency of automatic neural architecture design has been significantly improved by gradient-based search methods such as DARTS. However, recent literature has brought doubt to the generalization ability of DARTS, arguing that DARTS performs poorly when the search space is changed, i.e, when different s…
Quantum datasets improve QML performance.
Heteroskedasticity biases uplift model rankings, leading to inefficient treatment allocation.
This paper poses some basic questions about instances (hard to find) of a special problem in 3-manifold topology. "Important though the general concepts and propositions may be with the modern industrious passion for axiomatizing and generalizing has presented us ... nevertheless I am convinced that the special problem…
Authorship verification (AV) is a research subject in the field of digital text forensics that concerns itself with the question, whether two documents have been written by the same person. During the past two decades, an increasing number of proposed AV approaches can be observed. However, a closer look at the respect…
Financial time series forecasting is, without a doubt, the top choice of computational intelligence for finance researchers from both academia and financial industry due to its broad implementation areas and substantial impact. Machine Learning (ML) researchers came up with various models and a vast number of studies h…
During the past five years the Bayesian deep learning community has developed increasingly accurate and efficient approximate inference procedures that allow for Bayesian inference in deep neural networks. However, despite this algorithmic progress and the promise of improved uncertainty quantification and sample effic…
The study tests inferences about neural network optimization from linear interpolation of loss landscapes.
New findings show neural networks can be fooled by adversarial data, but a simple bias fix works.
There certainly is little or no doubt that politicians, sometimes consciously and sometimes not, exert a significant impact on stock markets. The evolving volatility over the Republican Donald Trump's surprise victory in the US presidential election is a perfect example when politicians, through announced policies, sen…