**Click to Read Paper and Get Code**

GeoSeq2Seq: Information Geometric Sequence-to-Sequence Networks

Jan 05, 2018

Alessandro Bay, Biswa Sengupta

Jan 05, 2018

Alessandro Bay, Biswa Sengupta

**Click to Read Paper and Get Code**

Approximating meta-heuristics with homotopic recurrent neural networks

Sep 07, 2017

Alessandro Bay, Biswa Sengupta

Sep 07, 2017

Alessandro Bay, Biswa Sengupta

**Click to Read Paper and Get Code**

Filtering Point Targets via Online Learning of Motion Models

Feb 20, 2019

Mehryar Emambakhsh, Alessandro Bay, Eduard Vazquez

Feb 20, 2019

Mehryar Emambakhsh, Alessandro Bay, Eduard Vazquez

* arXiv admin note: text overlap with arXiv:1806.06594

**Click to Read Paper and Get Code**

Convolutional Recurrent Predictor: Implicit Representation for Multi-target Filtering and Tracking

Nov 01, 2018

Mehryar Emambakhsh, Alessandro Bay, Eduard Vazquez

Nov 01, 2018

Mehryar Emambakhsh, Alessandro Bay, Eduard Vazquez

**Click to Read Paper and Get Code**

Deep Recurrent Neural Network for Multi-target Filtering

Oct 08, 2018

Mehryar Emambakhsh, Alessandro Bay, Eduard Vazquez

This paper addresses the problem of fixed motion and measurement models for multi-target filtering using an adaptive learning framework. This is performed by defining target tuples with random finite set terminology and utilisation of recurrent neural networks with a long short-term memory architecture. A novel data association algorithm compatible with the predicted tracklet tuples is proposed, enabling the update of occluded targets, in addition to assigning birth, survival and death of targets. The algorithm is evaluated over a commonly used filtering simulation scenario, with highly promising results.
Oct 08, 2018

Mehryar Emambakhsh, Alessandro Bay, Eduard Vazquez

* The 25th International Conference on MultiMedia Modeling (MMM)

**Click to Read Paper and Get Code**

Real-time tracker with fast recovery from target loss

Feb 12, 2019

Alessandro Bay, Panagiotis Sidiropoulos, Eduard Vazquez, Michele Sasdelli

Feb 12, 2019

Alessandro Bay, Panagiotis Sidiropoulos, Eduard Vazquez, Michele Sasdelli

* arXiv admin note: substantial text overlap with arXiv:1806.07844

**Click to Read Paper and Get Code**

Hide and Seek tracker: Real-time recovery from target loss

Jun 20, 2018

Alessandro Bay, Panagiotis Sidiropoulos, Eduard Vazquez, Michele Sasdelli

Jun 20, 2018

Alessandro Bay, Panagiotis Sidiropoulos, Eduard Vazquez, Michele Sasdelli

**Click to Read Paper and Get Code**

**Click to Read Paper and Get Code**

Emergence of Invariance and Disentanglement in Deep Representations

Jun 28, 2018

Alessandro Achille, Stefano Soatto

Jun 28, 2018

Alessandro Achille, Stefano Soatto

* Deep learning, neural network, representation, flat minima, information bottleneck, overfitting, generalization, sufficiency, minimality, sensitivity, information complexity, stochastic gradient descent, regularization, total correlation, PAC-Bayes

**Click to Read Paper and Get Code**

Credal Classification based on AODE and compression coefficients

Mar 27, 2012

Giorgio Corani, Alessandro Antonucci

Mar 27, 2012

Giorgio Corani, Alessandro Antonucci

**Click to Read Paper and Get Code**

Reliable Uncertain Evidence Modeling in Bayesian Networks by Credal Networks

Feb 15, 2018

Sabina Marchetti, Alessandro Antonucci

Feb 15, 2018

Sabina Marchetti, Alessandro Antonucci

* 19 pages

**Click to Read Paper and Get Code**

A Separation Principle for Control in the Age of Deep Learning

Nov 09, 2017

Alessandro Achille, Stefano Soatto

We review the problem of defining and inferring a "state" for a control system based on complex, high-dimensional, highly uncertain measurement streams such as videos. Such a state, or representation, should contain all and only the information needed for control, and discount nuisance variability in the data. It should also have finite complexity, ideally modulated depending on available resources. This representation is what we want to store in memory in lieu of the data, as it "separates" the control task from the measurement process. For the trivial case with no dynamics, a representation can be inferred by minimizing the Information Bottleneck Lagrangian in a function class realized by deep neural networks. The resulting representation has much higher dimension than the data, already in the millions, but it is smaller in the sense of information content, retaining only what is needed for the task. This process also yields representations that are invariant to nuisance factors and having maximally independent components. We extend these ideas to the dynamic case, where the representation is the posterior density of the task variable given the measurements up to the current time, which is in general much simpler than the prediction density maintained by the classical Bayesian filter. Again this can be finitely-parametrized using a deep neural network, and already some applications are beginning to emerge. No explicit assumption of Markovianity is needed; instead, complexity trades off approximation of an optimal representation, including the degree of Markovianity.
Nov 09, 2017

Alessandro Achille, Stefano Soatto

**Click to Read Paper and Get Code**

A Bayesian Approach to Sparse plus Low rank Network Identification

Sep 26, 2015

Mattia Zorzi, Alessandro Chiuso

Sep 26, 2015

Mattia Zorzi, Alessandro Chiuso

**Click to Read Paper and Get Code**

Identification of stable models via nonparametric prediction error methods

Jul 02, 2015

Diego Romeres, Gianluigi Pillonetto, Alessandro Chiuso

Jul 02, 2015

Diego Romeres, Gianluigi Pillonetto, Alessandro Chiuso

* number of pages = 6, number of figures = 3

**Click to Read Paper and Get Code**

A scaled gradient projection method for Bayesian learning in dynamical systems

Feb 02, 2015

Silvia Bonettini, Alessandro Chiuso, Marco Prato

Feb 02, 2015

Silvia Bonettini, Alessandro Chiuso, Marco Prato

* SIAM Journal on Scientific Computing 37 (2015), A1297-A1318

**Click to Read Paper and Get Code**

Bayesian and regularization approaches to multivariable linear system identification: the role of rank penalties

Sep 29, 2014

Giulia Prando, Alessandro Chiuso, Gianluigi Pillonetto

Sep 29, 2014

Giulia Prando, Alessandro Chiuso, Gianluigi Pillonetto

* to appear in IEEE Conference on Decision and Control, 2014

**Click to Read Paper and Get Code**

A probabilistic network for the diagnosis of acute cardiopulmonary diseases

May 12, 2018

Alessandro Magrini, Davide Luciani, Federico Mattia Stefanini

May 12, 2018

Alessandro Magrini, Davide Luciani, Federico Mattia Stefanini

* Biometrical Journal, 60(1), 174-195, January 2018

* The DOI of the article published after peer review was added. A technical detail was added in Section 3.2, Formula 8 (as a consequence, the ID of all the subsequent formulas result augmented by 1 with respect to the previous version). The prior standard deviation of the Gamma distribution in Table 4 was fixed (in the previous version, the prior variance was indicated, instead)

**Click to Read Paper and Get Code**

VFunc: a Deep Generative Model for Functions

Jul 11, 2018

Philip Bachman, Riashat Islam, Alessandro Sordoni, Zafarali Ahmed

Jul 11, 2018

Philip Bachman, Riashat Islam, Alessandro Sordoni, Zafarali Ahmed

* To be presented at the ICML 2018 workshop on Prediction and Generative Modeling in Reinforcement Learning

**Click to Read Paper and Get Code**