arXiv research

LAWN normalizes logits to improve deep network adaptability and generalization.

In this paper we consider the geodesic X-ray transform with attenuation coefficient as a combination of smooth complex function and 1-form. We show that attenuated X-ray transform applied to the pair of tensors is injective modulo the natural obstruction.

For a two-dimensional simple magnetic system, we study the attenuated magnetic ray transform $I_{A,Φ}$, with attenuation given by a unitary connection $A$ and a skew-Hermitian Higgs field $Φ$. We give a description for the range of $I_{A,Φ}$ acting on $\mathbb C^n$-valued tensor fields.

We show that the attenuated geodesic ray transform on two dimensional simple surfaces is injective. Moreover we give a stability estimate and develop a reconstruction procedure.

Logit regularization induces logit clustering, affecting classifier performance.

Improves adversarial robustness by constraining logits with a bounded function.

It has been shown in [Pa1] that on a simple, compact Riemannian 2-manifold the attenuated geodesic ray transform, with attenuation given by a connection and Higgs field, is injective on functions and 1-forms modulo the natural obstruction. Furthermore, the scattering relation determines the connection and Higgs field m…

For a compact Riemannian surface with boundary we study attenuated geodesic transform of functions and differential forms. We generalize several known results on uniqueness and stability of this transform dropping condition of absence of conjugate points.

Logit-GFN accelerates GFlowNets training by scaling logits based on temperature.

A new logit model derived from the Weibull manifold.

Positron Emission Tomography (PET) is a functional imaging modality widely used in neuroscience studies. To obtain meaningful quantitative results from PET images, attenuation correction is necessary during image reconstruction. For PET/MR hybrid systems, PET attenuation is challenging as Magnetic Resonance (MR) images…

We describe the range of the attenuated ray transform of a unitary connection on a simple surface acting on functions and 1-forms. We use this to determine the range of the ray transform acting on symmetric tensor fields.

Proposes SOVR loss to improve adversarial robustness by increasing logit margins.

LatentNN corrects neural network attenuation bias in astronomical data.

We study the problem of recovery both the attenuation $a$ and the source $f$ in the attenuated X-ray transform in the plane. We study the linearization as well. It turns out that there are natural Hamiltonian flow that determines which singularities we can recover. If the perturbations $δa$, $δf$ are supported in a com…

Framework predicts and prepares for rain-induced microwave link attenuation.

In this paper, we develop improved techniques for defending against adversarial examples at scale. First, we implement the state of the art version of adversarial training at unprecedented scale on ImageNet and investigate whether it remains effective in this setting - an important open scientific question (Athalye et …

Generative classifier derived from any discriminative classifier rejects illegal inputs.

Attenuation correction is an essential requirement of positron emission tomography (PET) image reconstruction to allow for accurate quantification. However, attenuation correction is particularly challenging for PET-MRI as neither PET nor magnetic resonance imaging (MRI) can directly image tissue attenuation properties…

Logit dynamics formula reveals self-regulation in softmax policy gradient methods.

Proposes a convex model for mixed logit to handle individual heterogeneity.

Recently, Kannan et al. [2018] proposed several logit regularization methods to improve the adversarial robustness of classifiers. We show that the computationally fast methods they propose - Clean Logit Pairing (CLP) and Logit Squeezing (LSQ) - just make the gradient-based optimization problem of crafting adversarial …

We present two range characterizations for the attenuated geodesic X-ray transform defined on pairs of functions and one-forms on simple surfaces. Such characterizations are based on first isolating the range over sums of functions and one-forms, then separating each sub-range in two ways, first by implicit conditions,…

MANO normalizes logits to estimate test accuracy without labels.

We derive explicit reconstruction formulas for the attenuated geodesic X-ray transform over functions and, in the case of non-vanishing attenuation, vector fields, on a class of simple Riemannian surfaces with boundary. These formulas partly rely on new explicit approaches to construct continuous right-inverses for bac…

New method uses low logit rank to simplify complex language models.

Logit models are usually applied when studying individual travel behavior, i.e., to predict travel mode choice and to gain behavioral insights on traveler preferences. Recently, some studies have applied machine learning to model travel mode choice and reported higher out-of-sample predictive accuracy than traditional …

Logit distance bounds representational similarity of models.

We consider neural network training, in applications in which there are many possible classes, but at test-time, the task is a binary classification task of determining whether the given example belongs to a specific class, where the class of interest can be different each time the classifier is applied. For instance, …

We show that for a simple surface with boundary the attenuated ray transform in the presence of a unitary connection and a skew-Hermitian Higgs field is injective modulo the natural obstruction for functions and vector fields. We also show that the connection and the Higgs field are uniquely determined by the scatterin…

Solves open problem on simple surfaces with novel twistor correspondence.

New algorithm reduces switching costs in multinomial logit bandit problems.

We evaluate the robustness of Adversarial Logit Pairing, a recently proposed defense against adversarial examples. We find that a network trained with Adversarial Logit Pairing achieves 0.6% accuracy in the threat model in which the defense is considered. We provide a brief overview of the defense and the threat models…

The paper reviews machine learning methods in PET imaging.

In this paper, a geometric function is introduced to reflect the attenuation speed of impact of one firm's default to its partner. If two firms are competitions (copartners), the default intensity of one firm will decrease (increase) abruptly when the other firm defaults. As time goes on, the impact will decrease gradu…

ELM improves neural model embeddings for long-tail learning.

This article deals with stability issues related to geodesic X-ray transforms, where an interplay between the (attenuation type) weight in the transform and the underlying geometry strongly impact whether the problem is stable or unstable. In the unstable case, we also explain what types of artifacts are expected in te…

New method distills cloud models into edge-friendly ones.

Paper tackles long-tailed labels in classification problems.

Deep learning classifiers are known to be vulnerable to adversarial examples. A recent paper presented at ICML 2019 proposed a statistical test detection method based on the observation that logits of noisy adversarial examples are biased toward the true class. The method is evaluated on CIFAR-10 dataset and is shown t…

The standard Gibbs sampler of Mixed Multinomial Logit (MMNL) models involves sampling from conditional densities of utility parameters using Metropolis-Hastings (MH) algorithm due to unavailability of conjugate prior for logit kernel. To address this non-conjugacy concern, we propose the application of Pólygamma data a…

The paper proposes an efficient method to scale Bayesian inference for mixed multinomial logit models to very large datasets.

Proposes a new model for context-dependent decision-making.

In the geophysical field, seismic noise attenuation has been considered as a critical and long-standing problem, especially for the pre-stack data processing. Here, we propose a model to leverage the deep-learning model for this task. Rather than directly applying an existing de-noising model from ordinary images to th…

New method estimates sparse covariance matrices in logit mixtures.

Inverts operator on hyperbolic surfaces, constructing invariant distributions.

Logit correction improves model performance by correcting spurious correlations.

This article presents a proof of the existence of Bertrand-Nash equilibrium prices with multi-product firms and under the Logit model of demand that does not rely on restrictive assumptions on product characteristics, firm homogeneity or symmetry, product costs, or linearity of the utility function. The proof is based …