Monotonic function ≈ Monotonic function
View article
Mish: A Self Regularized Non-Monotonic Neural Activation Function Open
The concept of non-linearity in a Neural Network is introduced by an activation function which serves an integral role in the training and performance evaluation of the network. Over the years of theoretical research, many activation funct…
View article
Regularizing and Optimizing LSTM Language Models Open
Recurrent neural networks (RNNs), such as long short-term memory networks (LSTMs), serve as a fundamental building block for many sequence learning tasks, including machine translation, language modeling, and question answering. In this pa…
View article
Monotonic Value Function Factorisation for Deep Multi-Agent Reinforcement Learning Open
In many real-world settings, a team of agents must coordinate its behaviour while acting in a decentralised fashion. At the same time, it is often possible to train the agents in a centralised fashion where global state information is avai…
View article
Wise-IoU: Bounding Box Regression Loss with Dynamic Focusing Mechanism Open
The loss function for bounding box regression (BBR) is essential to object detection. Its good definition will bring significant performance improvement to the model. Most existing works assume that the examples in the training data are hi…
View article
QMIX: Monotonic Value Function Factorisation for Deep Multi-Agent Reinforcement Learning Open
In many real-world settings, a team of agents must coordinate their behaviour while acting in a decentralised way. At the same time, it is often possible to train the agents in a centralised fashion in a simulated or laboratory setting, wh…
View article
Simulating masonry wall behaviour using a simplified micro-model approach Open
In this paper, a simplified micro-model approach utilising a combination of plasticity-based constitutive models and the extended finite element method (XFEM) is proposed. The approach is shown to be an efficient means of simulating the th…
View article
Trace-distance measure of coherence Open
We show that trace distance measure of coherence is a strong monotone for all qubit and, so called, $X$ states. An expression for the trace distance coherence for all pure states and a semi definite program for arbitrary states is provided…
View article
Neural Cognitive Diagnosis for Intelligent Education Systems Open
Cognitive diagnosis is a fundamental issue in intelligent education, which aims to discover the proficiency level of students on specific knowledge concepts. Existing approaches usually mine linear interactions of student exercising proces…
View article
An Optimal Iterative Learning Control Approach for Linear Systems With Nonuniform Trial Lengths Under Input Constraints Open
In practical applications of iterative learning control (ILC), the repetitive process may end up early by accident during the performance improvement along the trial axis, which yields the nonuniform trial length problem. For such practica…
View article
On the Properties of the Softmax Function with Application in Game Theory and Reinforcement Learning Open
In this paper, we utilize results from convex analysis and monotone operator theory to derive additional properties of the softmax function that have not yet been covered in the existing literature. In particular, we show that the softmax …
View article
A Forward-Backward Splitting Method for Monotone Inclusions Without Cocoercivity Open
In this work, we propose a simple modification of the forward-backward splitting method for finding a zero in the sum of two monotone operators. Our method converges under the same assumptions as Tseng's forward-backward-forward method, na…
View article
The Hypervolume Indicator Open
The hypervolume indicator is one of the most used set-quality indicators for the assessment of stochastic multiobjective optimizers, as well as for selection in evolutionary multiobjective optimization algorithms. Its theoretical propertie…
View article
Nonmonotonic Energy Dependence of Net-Proton Number Fluctuations Open
Nonmonotonic variation with collision energy (sqrt[s_{NN}]) of the moments of the net-baryon number distribution in heavy-ion collisions, related to the correlation length and the susceptibilities of the system, is suggested as a signature…
View article
Glow-TTS: A Generative Flow for Text-to-Speech via Monotonic Alignment Search Open
Recently, text-to-speech (TTS) models such as FastSpeech and ParaNet have been proposed to generate mel-spectrograms from text in parallel. Despite the advantage, the parallel TTS models cannot be trained without guidance from autoregressi…
View article
Online and Linear-Time Attention by Enforcing Monotonic Alignments Open
Recurrent neural network models with an attention mechanism have proven to be extremely effective on a wide variety of sequence-to-sequence problems. However, the fact that soft attention mechanisms perform a pass over the entire input seq…
View article
Cyclical Learning Rates for Training Neural Networks Open
It is known that the learning rate is the most important hyper-parameter to tune for training deep neural networks. This paper describes a new method for setting the learning rate, named cyclical learning rates, which practically eliminate…
View article
Quantum-coherence quantifiers based on the Tsallis relativeentropies Open
The concept of coherence is one of cornerstones in physics. The development\nof quantum information science has lead to renewed interest in properly\napproaching the coherence at the quantum level. Various measures could be\nproposed to qu…
View article
Modelling monotonic effects of ordinal predictors in Bayesian regression models Open
Ordinal predictors are commonly used in regression models. They are often incorrectly treated as either nominal or metric, thus under‐ or overestimating the information contained. Such practices may lead to worse inference and predictions …
View article
Symbolic Discovery of Optimization Algorithms Open
We present a method to formulate algorithm discovery as program search, and apply it to discover optimization algorithms for deep neural network training. We leverage efficient search techniques to explore an infinite and sparse program sp…
View article
Multi-Topic Misinformation Blocking With Budget Constraint on Online Social Networks Open
Along with the development of Information Technology, Online Social Networks (OSN) are constantly developing and have become popular media in the world. Besides communication enhancement benefits, OSN have such limitations on rapid spread …
View article
Zero-Reference Deep Curve Estimation for Low-Light Image Enhancement Open
The paper presents a novel method, Zero-Reference Deep Curve Estimation (Zero-DCE), which formulates light enhancement as a task of image-specific curve estimation with a deep network. Our method trains a lightweight deep network, DCE-Net,…
View article
On the Weak Convergence of the Extragradient Method for Solving Pseudo-Monotone Variational Inequalities Open
In infinite-dimensional Hilbert spaces, we prove that the iterative sequence generated by the extragradient method for solving pseudo-monotone variational inequalities converges weakly to a solution. A class of pseudo-monotone variational …
View article
Dynamical Systems Coupled with Monotone Set-Valued Operators: Formalisms, Applications, Well-Posedness, and Stability Open
International audience
View article
Towards Understanding Knowledge Distillation Open
Knowledge distillation, i.e., one classifier being trained on the outputs of another classifier, is an empirically very successful technique for knowledge transfer between classifiers. It has even been observed that classifiers learn much …
View article
Myths About Linear and Monotonic Associations: Pearson’s <i>r</i>, Spearman’s <i>ρ</i>, and Kendall’s <i>τ</i> Open
Pearson’s correlation coefficient is considered a measure of linear association between bivariate random variables X and Y. It is recommended not to use it for other forms of associations. Indeed, for nonlinear monotonic associations alter…
View article
Delay Compensation-Based State Estimation for Time-Varying Complex Networks With Incomplete Observations and Dynamical Bias Open
In this article, a delay-compensation-based state estimation (DCBSE) method is given for a class of discrete time-varying complex networks (DTVCNs) subject to network-induced incomplete observations (NIIOs) and dynamical bias. The NIIOs in…
View article
A general phase-field model for fatigue failure in brittle and ductile solids Open
In this work, the phase-field approach to fracture is extended to model fatigue failure in high- and low-cycle regime. The fracture energy degradation due to the repeated externally applied loads is introduced as a function of a local ener…
View article
When to Trust Your Model: Model-Based Policy Optimization Open
Designing effective model-based reinforcement learning algorithms is difficult because the ease of data generation must be weighed against the bias of model-generated data. In this paper, we study the role of model usage in policy optimiza…
View article
Mpemba Index and Anomalous Relaxation Open
The Mpemba effect is a counter-intuitive relaxation phenomenon, where a\nsystem prepared at a hot temperature cools down faster than an identical system\ninitiated at a cold temperature when both are quenched to an even colder bath.\nSuch …
View article
Morphological Inflection Generation with Hard Monotonic Attention Open
We present a neural model for morphological inflection generation which employs a hard attention mechanism, inspired by the nearly-monotonic alignment commonly found between the characters in a word and the characters in its inflection. We…