Francesco Dinuzzo

Francesco Dinuzzo

In this paper, we study the problem of identifying the impulse response of a linear time invariant (LTI) dynamical system from the knowledge of the input signal and a finite set of noisy output observations. We adopt an approach based on regularization in a Reproducing Kernel Hilbert Space (RKHS) that takes into account both continuous and discrete time systems. The focus of the paper is on designing spaces that are well suited for temporal impulse response modeling. To this end, we construct and characterize general families of kernels that incorporate system properties such as stability, relative degree, absence of oscillatory behavior, smoothness, or delay. In addition, we discuss the possibility of automatically searching over these classes by means of kernel learning techniques, so as to capture different modes of the system to be identified.

Garrett Wilson, Geoffrey Goh, Yan Jiang et al.

Detecting phishing, spam, fake accounts, data scraping, and other malicious activity in online social networks (OSNs) is a problem that has been studied for well over a decade, with a number of important results. Nearly all existing works on abuse detection have as their goal producing the best possible binary classifier; i.e., one that labels unseen examples as "benign" or "malicious" with high precision and recall. However, no prior published work considers what comes next: what does the service actually do after it detects abuse? In this paper, we argue that detection as described in previous work is not the goal of those who are fighting OSN abuse. Rather, we believe the goal to be selecting actions (e.g., ban the user, block the request, show a CAPTCHA, or "collect more evidence") that optimize a tradeoff between harm caused by abuse and impact on benign users. With this framing, we see that enlarging the set of possible actions allows us to move the Pareto frontier in a way that is unattainable by simply tuning the threshold of a binary classifier. To demonstrate the potential of our approach, we present Predictive Response Optimization (PRO), a system based on reinforcement learning that utilizes available contextual information to predict future abuse and user-experience metrics conditioned on each possible action, and select actions that optimize a multi-dimensional tradeoff between abuse/harm and impact on user experience. We deployed versions of PRO targeted at stopping automated activity on Instagram and Facebook. In both cases our experiments showed that PRO outperforms a baseline classification system, reducing abuse volume by 59% and 4.5% (respectively) with no negative impact to users. We also present several case studies that demonstrate how PRO can quickly and automatically adapt to changes in business constraints, system behavior, and/or adversarial tactics.

Jean-Baptiste Fiot, Francesco Dinuzzo

We explore the application of kernel-based multi-task learning techniques to forecast the demand of electricity in multiple nodes of a distribution network. We show that recently developed output kernel learning techniques are particularly well suited to solve this problem, as they allow to flexibly model the complex seasonal effects that characterize electricity demand data, while learning and exploiting correlations between multiple demand profiles. We also demonstrate that kernels with a multiplicative structure yield superior predictive performance with respect to the widely adopted (generalized) additive models. Our study is based on residential and industrial smart meter data provided by the Irish Commission for Energy Regulation (CER).

Francesco Dinuzzo

Simultaneously solving multiple related learning tasks is beneficial under a variety of circumstances, but the prior knowledge necessary to correctly model task relationships is rarely available in practice. In this paper, we develop a novel kernel-based multi-task learning technique that automatically reveals structural inter-task relationships. Building over the framework of output kernel learning (OKL), we introduce a method that jointly learns multiple functions and a low-rank multi-task kernel by solving a non-convex regularization problem. Optimization is carried out via a block coordinate descent strategy, where each subproblem is solved using suitable conjugate gradient (CG) type iterative methods for linear operator equations. The effectiveness of the proposed approach is demonstrated on pharmacological and collaborative filtering data.

Francesco Dinuzzo, Bernhard Schölkopf

A family of regularization functionals is said to admit a linear representer theorem if every member of the family admits minimizers that lie in a fixed finite dimensional subspace. A recent characterization states that a general class of regularization functionals with differentiable regularizer admits a linear representer theorem if and only if the regularization term is a non-decreasing function of the norm. In this report, we improve over such result by replacing the differentiability assumption with lower semi-continuity and deriving a proof that is independent of the dimensionality of the space.

Krikamol Muandet, Kenji Fukumizu, Francesco Dinuzzo et al.

This paper presents a kernel-based discriminative learning framework on probability measures. Rather than relying on large collections of vectorial training examples, our framework learns using a collection of probability distributions that have been constructed to meaningfully represent training data. By representing these probability distributions as mean embeddings in the reproducing kernel Hilbert space (RKHS), we are able to apply many standard kernel-based learning techniques in straightforward fashion. To accomplish this, we construct a generalization of the support vector machine (SVM) called a support measure machine (SMM). Our analyses of SMMs provides several insights into their relationship to traditional SVMs. Based on such insights, we propose a flexible SVM (Flex-SVM) that places different kernel functions on each training example. Experimental results on both synthetic and real-world data demonstrate the effectiveness of our proposed framework.