This paper defines less discriminatory algorithms and explores their feasibility.
arXiv research
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.
Trend · papers per month
Procedure for determining less discriminatory alternatives in AI audits with limited resources.
LDA-XGB1 balances fairness and accuracy in lending models.
Paper exposes vulnerabilities in interpreting machine learning models using adversarial attacks on PD plots.
The paper examines the stability of binary choice models using Gini index and scoring indicators.
We observe standard transfer learning can improve prediction accuracies of target tasks at the cost of lowering their prediction fairness -- a phenomenon we named discriminatory transfer. We examine prediction fairness of a standard hypothesis transfer algorithm and a standard multi-task learning algorithm, and show th…
The study compares uniform-price and discriminatory auctions in terms of learning difficulty.
The intention with this paper is to provide all the estimation concepts and techniques that are needed to implement a two-phases approach to the parametric estimation of probability of default (PD) curves. In the first phase of this approach, a raw PD curve is estimated based on parameters that reflect discriminatory p…
Private release of sensitive data enables fair learning.
Point estimation of class prevalences in the presence of data set shift has been a popular research topic for more than two decades. Less attention has been paid to the construction of confidence and prediction intervals for estimates of class prevalences. One little considered question is whether or not it is necessar…
We prove that the degree of the Brandt-Lickorish-Millet polynomial of any quasi-alternating link is less than its determinant. Therefore, we obtain a new and a simple obstruction criterion for quasi-alternateness. As an application, we identify some knots of 12 crossings or less and some links of 9 crossings or less th…
We show that the crossing number of any link that is known to be quasi-alternating is less than or equal to its determinant. Based on this, we conjecture that the crossing number of any quasi-alternating link is less than or equal to its determinant. Thus, if this conjecture is proved then it would give an easier obstr…
r-STSF improves TSC accuracy and interpretability.
This paper elaborates on the validation requirements for rating systems and probabilities of default (PDs) which were introduced with the New Capital Standards (Basel II). We start in Section 2 with some introductory remarks on the topics and approaches that will be discussed later on. Then we have a view on the develo…
Qazaqzeh and Chbili showed that for any quasi-alternating link, the degree of -polynomial is less than its determinant. We give a refinement of their evaluation.
Study on Fox's trapezoidal conjecture for specific alternating links.
Optimal projections enhance Naive Bayes classification.
Explicit isometry groups found for nearly Kähler manifolds.
Recent research has helped to cultivate growing awareness that machine learning systems fueled by big data can create or exacerbate troubling disparities in society. Much of this research comes from outside of the practicing data science community, leaving its members with little concrete guidance to proactively addres…
A classical result states that the determinant of an alternating link is equal to the number of spanning trees in a checkerboard graph of an alternating connected projection of the link. We generalize this result to show that the determinant is the alternating sum of the number of quasi-trees of genus j of the dessin o…
A machine learning model may exhibit discrimination when used to make decisions involving people. One potential cause for such outcomes is that the model uses a statistical proxy for a protected demographic attribute. In this paper we formulate a definition of proxy use for the setting of linear regression and present …
Constructs links that are both quasi-alternating and almost alternating.
We calculate the alternating number of torus knots with braid index 4 and less. For the lower bound, we use the upsilon-invariant recently introduced by Ozsváth, Stipsicz, and Szabó. For the upper bound, we use a known bound for braid index and a new bound for braid index . Both bounds coincide, so that we obtai…
With the industry trend of shifting from a traditional hierarchical approach to flatter management structure, crowdsourced performance assessment gained mainstream popularity. One fundamental challenge of crowdsourced performance assessment is the risks that personal interest can introduce distortions of facts, especia…
New metric MADD assesses fairness of predictive student models.
Expands Bayesian experiment design framework to account for model discrepancies.
Many datasets can be viewed as a noisy sampling of an underlying space, and tools from topological data analysis can characterize this structure for the purpose of knowledge discovery. One such tool is persistent homology, which provides a multiscale description of the homological features within a dataset. A useful re…
Efficient methods estimate concordance probability for big data.
Automated decision making systems are increasingly being used in real-world applications. In these systems for the most part, the decision rules are derived by minimizing the training error on the available historical data. Therefore, if there is a bias related to a sensitive attribute such as gender, race, religion, e…
Accuracies of survival models for life expectancy prediction as well as critical-care applications are significantly compromised due to the sparsity of samples and extreme imbalance between the survival (usually, the majority) and mortality class sizes. While a recent random survival forest (RSF) model overcomes the li…
We use Heegaard Floer homology to give obstructions to unknotting a knot with a single crossing change. These restrictions are particularly useful in the case where the knot in question is alternating. As an example, we use them to classify all knots with crossing number less than or equal to nine and unknotting number…
It is known that the arc index of alternating knots is the minimal crossing number plus two and the arc index of prime nonalternating knots is less than or equal to the minimal crossing number. We study some cases when the arc index is strictly less than the minimal crossing number. We also give minimal grid diagrams o…
Discriminatory trade liberalization policies are becoming more popular among world economies. Countries are motivated to enter for regional trade agreements to capture faster economic growth for alleviating poverty. In developing economies like most of the member countries of the Association of South East Asian Nations…
Data-driven algorithms play a large role in decision making across a variety of industries. Increasingly, these algorithms are being used to make decisions that have significant ramifications for people's social and economic well-being, e.g. in sentencing, loan approval, and policing. Amid the proliferation of such sys…
As algorithms increasingly inform and influence decisions made about individuals, it becomes increasingly important to address concerns that these algorithms might be discriminatory. The output of an algorithm can be discriminatory for many reasons, most notably: (1) the data used to train the algorithm might be biased…
Bayesian optimization (BO) and its batch extensions are successful for optimizing expensive black-box functions. However, these traditional BO approaches are not yet ideal for optimizing less expensive functions when the computational cost of BO can dominate the cost of evaluating the blackbox function. Examples of the…
Study data biases to predict algorithmic discrimination, developing a Data Bias Profile.
Whole MILC learns brain disorder dynamics from unlabeled data.
We say that a link is an s-major of a link if any diagram of can be transformed into a diagram of by changing some crossings and smoothing some crossings. This relation is a partial ordering on the set of all prime alternating links. We determine this partial order for all prime alternating knot…
Games generalize the single-objective optimization paradigm by introducing different objective functions for different players. Differentiable games often proceed by simultaneous or alternating gradient updates. In machine learning, games are gaining new importance through formulations like generative adversarial netwo…
Alternating links bound rational homology balls if their chessboard lattice is cubiquitous.
We extend an approach of Beliakova for computing knot Floer homology and implement it in a publicly available computer program. We review the main programming and optimization methods used. Our program is then used to check that the Floer homology of a prime non-alternating knot with less than 12 crossings has no torsi…
Paper tackles low-rank matrix recovery with column -norm regularization.
The paper addresses the -tangle enumeration problem. We introduce a notion of cascade diagram for -tangle projections. An effective enumeration algorithm for projections is proposed based on cascade representation. Tangles projections with up to 12 crossings are tabulated. We provide also pictures of alternating …
This paper computes Kakimizu complexes for all 11 crossing prime alternating knots.
We give an alternative proof of Madsen-Weiss' generalized Mumford conjecture. Our proof is based on ideas similar to Madsen-Weiss' original proof, but it is more geometrical and less homotopy theoretical in nature. At the heart of the argument is a geometric version of Harer stability, which we formulate as a theorem a…
It is well known that the braid index of a link equals the minimum number of Seifert circles among all link diagrams representing it. For a link with a reduced alternating diagram , , the number of Seifert circles in , equals the braid index of if contains no {\em lone crossings} (a …
We explore the use of graph networks to deal with irregular-geometry detectors in the context of particle reconstruction. Thanks to their representation-learning capabilities, graph networks can exploit the full detector granularity, while natively managing the event sparsity and arbitrarily complex detector geometries…