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.
VALC provides concept-level interpretations of FLMs, overcoming word-level limitations.
PACE explains ViTs by modeling patch-level concept distributions, surpassing existing methods.
Simplifies F-measure for better interpretability.
The last decade has seen huge progress in the development of advanced machine learning models; however, those models are powerless unless human users can interpret them. Here we show how the mind's construction of concepts and meaning can be used to create more interpretable machine learning models. By proposing a nove…
Geometric framework detects concept frustration between human concepts and machine representations.
Concept Hierarchies and Formal Concept Analysis are theoretically well grounded and largely experimented methods. They rely on line diagrams called Galois lattices for visualizing and analysing object-attribute sets. Galois lattices are visually seducing and conceptually rich for experts. However they present important…
TRUST improves tree models' accuracy while maintaining interpretability.
The paper introduces metrics to objectively evaluate interpretability methods.
We consider the problem of estimating a regression function in the common situation where the number of features is small, where interpretability of the model is a high priority, and where simple linear or additive models fail to provide adequate performance. To address this problem, we present GapTV, an approach that …
A new framework quantifies how model explanations influence each other.
We propose a conceptual design for a quantum blockchain. Our method involves encoding the blockchain into a temporal GHZ (Greenberger-Horne-Zeilinger) state of photons that do not simultaneously coexist. It is shown that the entanglement in time, as opposed to an entanglement in space, provides the crucial quantum adva…
We consider the problem of estimating a regression function in the common situation where the number of features is small, where interpretability of the model is a high priority, and where simple linear or additive models fail to provide adequate performance. To address this problem, we present Maximum Variance Total V…
To study how mental object representations are related to behavior, we estimated sparse, non-negative representations of objects using human behavioral judgments on images representative of 1,854 object categories. These representations predicted a latent similarity structure between objects, which captured most of the…
We revisit the linearization theorems for proper Lie groupoids around general orbits (statements and proofs). In the the fixed point case (known as Zung's theorem) we give a shorter and more geometric proof, based on a Moser deformation argument. The passing to general orbits (Weinstein) is given a more conceptual inte…
Regularizes GAMs to improve interpretability by reducing concurvity.
The study designs inherently interpretable machine learning models for high-risk sectors.
Analogous zeta function for twisted Alexander invariants defined.
We formulate a more conceptual interpretation of the Cappell-Lee-Miller glueing/splitting theorem using the new language of asymptotic maps and asymptotic exactness. Additionally, we present an asymptotic description of the Mayer-Vietoris sequence naturally associated to the Cech cohomology of the sheaf of local soluti…
We note that the Bogomolny equation for abelian vortices is precisely the condition for invariance of the Hermitian-Einstein equation under a degenerate conformal transformation. This leads to a natural interpretation of vortices as degenerate hermitian metrics that satisfy a certain curvature equation. Using this view…
We introduce two tools, dynamical thickening and flow selectors, to overcome the infamous discontinuity of the gradient flow endpoint map near non-degenerate critical points. More precisely, we interpret the stable fibrations of certain Conley pairs , established in [2,3], as a dynamical thickening of the stable…
Novel framework for risk-sensitive reinforcement learning using martingale decomposition.
APD method decomposes neural network parameters into simple, faithful components.
New model explains how concepts grow based on experience.
New model combines neural networks and embeddings for better choice modeling interpretability.
New proof of chain duality for simplicial complexes.
AI mirrors modern math's autonomous development, raising interpretive challenges.
The cognitive framework of conceptual spaces bridges the gap between symbolic and subsymbolic AI by proposing an intermediate conceptual layer where knowledge is represented geometrically. There are two main approaches for obtaining the dimensions of this conceptual similarity space: using similarity ratings from psych…
New framework models uncertainty in classification debates.
This paper presents a computational model for conceptual shifts, based on a novelty metric applied to a vector representation generated through deep learning. This model is integrated into a co-creative design system, which enables a partnership between an AI agent and a human designer interacting through a sketching c…
We introduce a simple recurrent variational auto-encoder architecture that significantly improves image modeling. The system represents the state-of-the-art in latent variable models for both the ImageNet and Omniglot datasets. We show that it naturally separates global conceptual information from lower level details, …
Causal Bayesian networks interpret actions as interventions to connect models to real-world outcomes.
GeoShapley uses game theory to measure spatial effects in ML models.
Algorithm improves SLR efficiency in financial narratives.
Following the programme set out in Part I of this work, we develop a conceptual higher order differential calculus. The '' local linear algebra '' defined in Part I is generalized by '' higher order local linear algebra ''. The underlying combinatorial object of such higher algebra is the natural n-dimensional hyper-cu…
Deep networks can be understood as logical circuits, improving interpretability and generalization.
New methods identify concepts in trained embeddings reliably without human labels.
We map the recently proposed notions of algorithmic fairness to economic models of Equality of opportunity (EOP)---an extensively studied ideal of fairness in political philosophy. We formally show that through our conceptual mapping, many existing definition of algorithmic fairness, such as predictive value parity and…
NDM incorporates geometric structure into neural networks for better optimization and interpretability.
Calibration without labels in multiple testing
LFD method improves text classification by making features clearer and less label-leaking.
Online news media provides aggregated news and stories from different sources all over the world and up-to-date news coverage. The main goal of this study is to have a solution that considered as a homogeneous source for the news and to represent the news in a new conceptual framework. Furthermore, the user can easily …
The paper extends static Systemic Risk Measures to a conditional setting.
Artificial intelligence (AI) is intrinsically data-driven. It calls for the application of statistical concepts through human-machine collaboration during generation of data, development of algorithms, and evaluation of results. This paper discusses how such human-machine collaboration can be approached through the sta…
This work abstracts deep neural networks into concept graphs for better interpretability in medical tasks.
The paper introduces GAER to assess market feasibility under geopolitical and institutional constraints.
We define invariants for a framed link equipped with a SL2 local system in its complement and additional combinatorial data based on the theory of representations of stated skein algebras at roots of unity of punctured bigons and the geometric interpretation of their centers. The gauge invariance of the link invariant …
We propose a new class of probabilistic neural-symbolic models, that have symbolic functional programs as a latent, stochastic variable. Instantiated in the context of visual question answering, our probabilistic formulation offers two key conceptual advantages over prior neural-symbolic models for VQA. Firstly, the pr…
Counterfactuals help compare different scenarios in CI and XAI.