Dropout is a very effective method in preventing overfitting and has become the go-to regularizer for multi-layer neural networks in recent years. Hierarchical mixture of experts is a hierarchically gated model that defines a soft decision tree where leaves correspond to experts and decision nodes correspond to gating models that softly choose between its children, and as such, the model defines a soft hierarchical partitioning of the input space. In this work, we propose a variant of dropout for hierarchical mixture of experts that is faithful to the tree hierarchy defined by the model, as opposed to having a flat, unitwise independent application of dropout as one has with multi-layer perceptrons. We show that on a synthetic regression data and on MNIST and CIFAR-10 datasets, our proposed dropout mechanism prevents overfitting on trees with many levels improving generalization and providing smoother fits.

Recommender systems have recently attracted many researchers in the deep learning community. The state-of-the-art deep neural network models used in recommender systems are typically multilayer perceptron and deep Autoencoder (DAE), among which DAE usually shows better performance due to its superior capability to reconstruct the inputs. However, we found existing DAE recommendation systems that have similar implementations on similar datasets result in vastly different parameter settings. In this work, we have built a flexible DAE model, named FlexEncoder that uses configurable parameters and unique features to analyse the parameter influences on the prediction accuracy of recommender systems. This will help us identify the best-performance parameters given a dataset. Extensive evaluation on the MovieLens datasets are conducted, which drives our conclusions on the influences of DAE parameters. Specifically, we find that DAE parameters strongly affect the prediction accuracy of the recommender systems, and the effect is transferable to similar datasets in a larger size. We open our code to public which could benefit both new users for DAE -- they can quickly understand how DAE works for recommendation systems, and experienced DAE users -- it easier for them to tune the parameters on different datasets.

Federated learning is an emerging technique used to prevent the leakage of private information. Unlike centralized learning that needs to collect data from users and store them collectively on a cloud server, federated learning makes it possible to learn a global model while the data are distributed on the users' devices. However, compared with the traditional centralized approach, the federated setting consumes considerable communication resources of the clients, which is indispensable for updating global models and prevents this technique from being widely used. In this paper, we aim to optimize the structure of the neural network models in federated learning using a multi-objective evolutionary algorithm to simultaneously minimize the communication costs and the global model test errors. A scalable method for encoding network connectivity is adapted to federated learning to enhance the efficiency in evolving deep neural networks. Experimental results on both multilayer perceptrons and convolutional neural networks indicate that the proposed optimization method is able to find optimized neural network models that can not only significantly reduce communication costs but also improve the learning performance of federated learning compared with the standard fully connected neural networks.

We study the problem of jointly reasoning about language and vision through a navigation and spatial reasoning task. We introduce the Touchdown task and dataset, where an agent must first follow navigation instructions in a real-life visual urban environment to a goal position, and then identify in the observed image a location described in natural language to find a hidden object. The data contains 9,326 examples of English instructions and spatial descriptions paired with demonstrations. We perform qualitative linguistic analysis, and show that the data displays richer use of spatial reasoning compared to related resources. Empirical analysis shows the data presents an open challenge to existing methods.

Despite the growing popularity of modern machine learning techniques (e.g. Deep Neural Networks) in cyber-security applications, most of these models are perceived as a black-box for the user. Adversarial machine learning offers an approach to increase our understanding of these models. In this paper we present an approach to generate explanations for incorrect classifications made by data-driven Intrusion Detection Systems (IDSs). An adversarial approach is used to find the minimum modifications (of the input features) required to correctly classify a given set of misclassified samples. The magnitude of such modifications is used to visualize the most relevant features that explain the reason for the misclassification. The presented methodology generated satisfactory explanations that describe the reasoning behind the mis-classifications, with descriptions that match expert knowledge. The advantages of the presented methodology are: 1) applicable to any classifier with defined gradients. 2) does not require any modification of the classifier model. 3) can be extended to perform further diagnosis (e.g. vulnerability assessment) and gain further understanding of the system. Experimental evaluation was conducted on the NSL-KDD99 benchmark dataset using Linear and Multilayer perceptron classifiers. The results are shown using intuitive visualizations in order to improve the interpretability of the results.

This technical report describes the multi-label classification (MLC) search space in the MEKA software, including the traditional/meta MLC algorithms, and the traditional/meta/pre-processing single-label classification (SLC) algorithms. The SLC search space is also studied because is part of MLC search space as several methods use problem transformation methods to create a solution (i.e., a classifier) for a MLC problem. This was done in order to understand better the MLC algorithms. Finally, we propose a grammar that formally expresses this understatement.

Italian researchers recently developed the first functioning quantum neural network by running a special algorithm on an actual quantum computer. The team, lead by Francesco Tacchino of the University of Pavia in Italy, pre-published their research on ArXiv earlier this month in a research paper titled "An Artificial Neuron Implemented on an Actual Quantum Processor." Basically, they developed a single-layer artificial neural network (ANN) that runs on a quantum computer. This kind of rudimentary ANN is called a perceptron, and it's the basic building block of more robust neural networks. Previous attempts at building a perceptron on a quantum system have involved treating individual qubits as neurons in a network.

In machine learning, the perceptron is an algorithm for supervised learning of binary classifiers. A binary classifier is a function which can decide whether or not an input, represented by a vector of numbers, belongs to some specific class. It is a type of linear classifier, i.e. a classification algorithm that makes its predictions based on a linear predictor function combining a set of weights with the feature vector. In simple words, Perceptron is a Computational Unit (or simply a function) which takes in set of inputs and gives out the result. The role of perceptron is to find a function which will approximately satisfy all the given data. Lets say X₁ (x₁1, x₁2, x₁3) are one of our data in a set of data X (X₁, X₂, …, Xₙ).

In the context of this course, we view neural networks as "just" another nonlinear hypothesis space. On the practical side, unlike trees and tree-based ensembles (our other major nonlinear hypothesis spaces), neural networks can be fit using gradient-based optimization methods. We discuss the specific case of the multilayer perceptron for multiclass classification, which we view as a generalization of multinomial logistic regression from linear to nonlinear score functions.

Breast cancer is the most frequently reported cancer type among the women around the globe and beyond that it has the second highest female fatality rate among all cancer types. Despite all the progresses made in prevention and early intervention, early prognosis and survival prediction rates are still unsatisfactory. In this paper, we propose a novel type of perceptron called L-Perceptron which outperforms all the previous supervised learning methods by reaching 97.42 \% and 98.73 \% in terms of accuracy and sensitivity, respectively in Wisconsin Breast Cancer dataset. Experimental results on Haberman's Breast Cancer Survival dataset, show the superiority of proposed method by reaching 75.18 \% and 83.86 \% in terms of accuracy and F1 score, respectively. The results are the best reported ones obtained in 10-fold cross validation in absence of any preprocessing or feature selection.