arXiv research

DeepCTRL integrates rules into deep learning models, allowing flexible control at inference.

In the artificial intelligence field, learning often corresponds to changing the parameters of a parameterized function. A learning rule is an algorithm or mathematical expression that specifies precisely how the parameters should be changed. When creating an artificial intelligence system, we must make two decisions: …

A major goal of unsupervised learning is to discover data representations that are useful for subsequent tasks, without access to supervised labels during training. Typically, this involves minimizing a surrogate objective, such as the negative log likelihood of a generative model, with the hope that representations us…

ProSMIN improves representation quality through probabilistic self-supervised learning.

Study explores reinforcement learning in a complex game environment, analyzing rule inference and policy learning.

We establish relationships between two classes of invariants of Legendrian knots in $\mathbb{R}^3$: Representation numbers of the Chekanov-Eliashberg DGA and satellite ruling polynomials. For positive permutation braids, $β\subset J^1S^1$, we give a precise formula in terms of representation numbers for the $m$-graded …

Homotopy cardinality counts augmentations of Legendrian knots.

Classifies surfaces with zero mean curvature in a light cone.

We present a framework for autonomously learning a portable representation that describes a collection of low-level continuous environments. We show that these abstract representations can be learned in a task-independent egocentric space specific to the agent that, when grounded with problem-specific information, are …

A central goal of meta-learning is to find a learning rule that enables fast adaptation across a set of tasks, by learning the appropriate inductive bias for that set. Most meta-learning algorithms try to find a \textit{global} learning rule that encodes this inductive bias. However, a global learning rule represented …

This paper considers exponential utility indifference pricing for a multidimensional non-traded assets model, and provides two linear approximations for the utility indifference price. The key tool is a probabilistic representation for the utility indifference price by the solution of a functional differential equation…

ASTRA uses unlabeled data and weak rules to train deep models effectively.

Machine learning accelerates Lie algebra computations.

For any Legendrian knot $K$ in standard contact ${\mathbb R}^3$ we relate counts of ungraded ($1$-graded) representations of the Legendrian contact homology DG-algebra $(\mathcal{A}(K),\partial)$ with the $n$-colored Kauffman polynomial. To do this, we introduce an ungraded $n$-colored ruling polynomial, $R^1_{n,K}(q)$…

As modern deep networks become more complex, and get closer to human-like capabilities in certain domains, the question arises of how the representations and decision rules they learn compare to the ones in humans. In this work, we study representations of sentences in one such artificial system for natural language pr…

The paper introduces new measures for quantifying uncertainty in machine learning.

New insights into state representations in RL help design better learning rules.

Analysis of deep neural networks under various learning rules reveals dynamics of feature and prediction learning.

We study satellites of Legendrian knots in R^3 and their relation to the Chekanov-Eliashberg differential graded algebra of the knot. In particular, we generalize the well-known correspondence between rulings of a Legendrian knot in R^3 and augmentations of its DGA by showing that the DGA has finite-dimensional represe…

We consider the problem of structure learning for Gaifman models and learn relational features that can be used to derive feature representations from a knowledge base. These relational features are first-order rules that are then partially grounded and counted over local neighborhoods of a Gaifman model to obtain the …

Paper learns data-driven organ matching rules from observational data.

Autonomous driving decision-making is a great challenge due to the complexity and uncertainty of the traffic environment. Combined with the rule-based constraints, a Deep Q-Network (DQN) based method is applied for autonomous driving lane change decision-making task in this study. Through the combination of high-level …

We study the representation theory of the quantum Teichmueller space when going to infinity in the classical Teichmueller space. The geometric ingredients are the extension of Thurston's shear coordinates to the augmented Teichmueller space and the study of the Weil-Petersson Poisson structure for this extension. The r…

Humans navigate complex environments in an organized yet flexible manner, adapting to the context and implicit social rules. Understanding these naturally learned patterns of behavior is essential for applications such as autonomous vehicles. However, algorithmically defining these implicit rules of human behavior rema…

We use categorical skew Howe duality to find recursion rules that compute categorified sl(N) invariants of rational tangles colored by exterior powers of the standard representation. Further, we offer a geometric interpretation of these rules which suggests a connection to Floer theory. Along the way we make progress t…

Derives integral representations for a Lévy process and its extremum, hitting time, with fast evaluation.

TransINT embeds KGs by preserving implication rules, outperforming existing methods.

Unified quadrature framework for large-scale kernel machines.

A nonparametric kernel-based method for realizing Bayes' rule is proposed, based on representations of probabilities in reproducing kernel Hilbert spaces. Probabilities are uniquely characterized by the mean of the canonical map to the RKHS. The prior and conditional probabilities are expressed in terms of RKHS functio…

This note is a write-up of a talk given by the author at the Meeting of the Sociedade Portuguesa de Matematica in July 2012. We describe Jaeger's HOMFLY-PT expansion of the Kauffman polynomial and how to generalize it to other quantum invariants using the so-called "branching rules" for Lie algebra representations. We …

Study geometric bases for A-polynomials in SU(3) using arcade formalism.

Lasso is a widely used regression technique to find sparse representations. When the dimension of the feature space and the number of samples are extremely large, solving the Lasso problem remains challenging. To improve the efficiency of solving large-scale Lasso problems, El Ghaoui and his colleagues have proposed th…

Study on pricing rules for income streams with partial insider information.

To deepen our understanding of graph neural networks, we investigate the representation power of Graph Convolutional Networks (GCN) through the looking glass of graph moments, a key property of graph topology encoding path of various lengths. We find that GCNs are rather restrictive in learning graph moments. Without c…

New Fourier features improve high-precision approximation in large-scale problems.

With the growing adoption of machine learning techniques, there is a surge of research interest towards making machine learning systems more transparent and interpretable. Various visualizations have been developed to help model developers understand, diagnose, and refine machine learning models. However, a large numbe…

Conventional techniques for supervised classification constrain the classification rules considered and use surrogate losses for classification 0-1 loss. Favored families of classification rules are those that enjoy parametric representations suitable for surrogate loss minimization, and low complexity properties suita…

New decision-theoretic characterization separates belief and decision posteriors.

This paper simplifies complex game dynamics by using a recursive representation.

In a physical neural system, where storage and processing are intimately intertwined, the rules for adjusting the synaptic weights can only depend on variables that are available locally, such as the activity of the pre- and post-synaptic neurons, resulting in local learning rules. A systematic framework for studying t…

DCR improves interpretability of concept-based models by using neural networks to build rule structures.

Different directed acyclic graphs (DAGs) may be Markov equivalent in the sense that they entail the same conditional independence relations among the observed variables. Meek (1995) characterizes Markov equivalence classes for DAGs (with no latent variables) by presenting a set of orientation rules that can correctly i…

A new concept of confidence in learning is defined and analyzed.

A test assesses the calibration of set-based epistemic uncertainty representations.

Paper refines InfoNCE for accurate mutual information estimation.

The detection of fraud in accounting data is a long-standing challenge in financial statement audits. Nowadays, the majority of applied techniques refer to handcrafted rules derived from known fraud scenarios. While fairly successful, these rules exhibit the drawback that they often fail to generalize beyond known frau…

Method measures weight similarity in neural networks using normalization and statistical inference.

We develop a functional encoder-decoder approach to supervised meta-learning, where labeled data is encoded into an infinite-dimensional functional representation rather than a finite-dimensional one. Furthermore, rather than directly producing the representation, we learn a neural update rule resembling functional gra…