Alessandro Rozza
YOU?
Author Swipe
View article: A Primal-Dual Online Learning Approach for Dynamic Pricing of Sequentially Displayed Complementary Items under Sale Constraints
A Primal-Dual Online Learning Approach for Dynamic Pricing of Sequentially Displayed Complementary Items under Sale Constraints Open
We address the challenging problem of dynamically pricing complementary items that are sequentially displayed to customers. An illustrative example is the online sale of flight tickets, where customers navigate through multiple web pages. …
View article: Maximum Entropy Exploration in Contextual Bandits with Neural Networks and Energy Based Models
Maximum Entropy Exploration in Contextual Bandits with Neural Networks and Energy Based Models Open
Contextual bandits can solve a huge range of real-world problems. However, current popular algorithms to solve them either rely on linear models or unreliable uncertainty estimation in non-linear models, which are required to deal with the…
View article: A survey and taxonomy of loss functions in machine learning
A survey and taxonomy of loss functions in machine learning Open
Most state-of-the-art machine learning techniques revolve around the optimisation of loss functions. Defining appropriate loss functions is therefore critical to successfully solving problems in this field. In this survey, we present a com…
View article: Maximum entropy exploration in contextual bandits with neural networks and energy based models
Maximum entropy exploration in contextual bandits with neural networks and energy based models Open
Contextual bandits can solve a huge range of real-world problems. However, current popular algorithms to solve them either rely on linear models, or unreliable uncertainty estimation in non-linear models, which are required to deal with th…
View article: Composition and Style Attributes Guided Image Aesthetic Assessment
Composition and Style Attributes Guided Image Aesthetic Assessment Open
The aesthetic quality of an image is defined as the measure or appreciation of the beauty of an image. Aesthetics is inherently a subjective property but there are certain factors that influence it such as, the semantic content of the imag…
View article: Ranking Micro-Influencers: a Novel Multi-Task Learning and Interpretable Framework
Ranking Micro-Influencers: a Novel Multi-Task Learning and Interpretable Framework Open
With the rise in use of social media to promote branded products, the demand for effective influencer marketing has increased. Brands are looking for improved ways to identify valuable influencers among a vast catalogue; this is even more …
View article: Modeling image aesthetics through aesthetics-related attributes
Modeling image aesthetics through aesthetics-related attributes Open
Automatic assessment of image aesthetics is a challenging task for the computer vision community that has a wide range of applications. The most promising state-of-the-art approaches are based on deep learning methods that jointly predict …
View article: No Reference, Opinion Unaware Image Quality Assessment by Anomaly Detection
No Reference, Opinion Unaware Image Quality Assessment by Anomaly Detection Open
We propose an anomaly detection based image quality assessment method which exploits the correlations between feature maps from a pre-trained Convolutional Neural Network (CNN). The proposed method encodes the intra-layer correlation throu…
View article: DOPSIE: Deep-Order Proximity and Structural Information Embedding
DOPSIE: Deep-Order Proximity and Structural Information Embedding Open
Graph-embedding algorithms map a graph into a vector space with the aim of preserving its structure and its intrinsic properties. Unfortunately, many of them are not able to encode the neighborhood information of the nodes well, especially…
View article: Automated Pruning for Deep Neural Network Compression
Automated Pruning for Deep Neural Network Compression Open
In this work we present a method to improve the pruning step of the current\nstate-of-the-art methodology to compress neural networks. The novelty of the\nproposed pruning technique is in its differentiability, which allows pruning to\nbe …
View article: Learning Combinations of Activation Functions
Learning Combinations of Activation Functions Open
In the last decade, an active area of research has been devoted to design\nnovel activation functions that are able to help deep neural networks to\nconverge, obtaining better performance. The training procedure of these\narchitectures usu…
View article: Making Deep Neural Networks Robust to Label Noise: a Loss Correction Approach
Making Deep Neural Networks Robust to Label Noise: a Loss Correction Approach Open
We present a theoretically grounded approach to train deep neural networks, including recurrent networks, subject to class-dependent label noise. We propose two procedures for loss correction that are agnostic to both application domain an…
View article: A Novel Mutual Information-based Feature Selection Algorithm.
A Novel Mutual Information-based Feature Selection Algorithm. Open
From a machine learning point of view to identify a subset of relevant features from a real data set can be useful to improve the results achieved by classification methods and to reduce their time and space complexity. To achieve this goa…
View article: A Cross-Entropy-based Method to Perform Information-based Feature Selection
A Cross-Entropy-based Method to Perform Information-based Feature Selection Open
From a machine learning point of view, identifying a subset of relevant features from a real data set can be useful to improve the results achieved by classification methods and to reduce their time and space complexity. To achieve this go…
View article: Intrinsic Dimension Estimation: Relevant Techniques and a Benchmark Framework
Intrinsic Dimension Estimation: Relevant Techniques and a Benchmark Framework Open
When dealing with datasets comprising high-dimensional points, it is usually advantageous to discover some data structure. A fundamental information needed to this aim is the minimum number of parameters required to describe the data while…