Archive

We'll discuss how Priberam is applying machine learning to the problems of Media Monitoring and Technology Watch in the context of the INSIGHT P2020 project. We'll show a working prototype and describe the developed microservices platform, and how the several NLP components come into play. In particular, we'll give an overview of the inner workings of the core machine learning models developed by Priberam, such...

Nowadays we are collecting high-dimensional and large data streams, where many dimensions can be expressing basically the same information on the underlying process of interest. This redundancy is apparent, for example, if we observe mass media news outputs through time. Here, even the discovery of the leader-follower structure of the news streams is valuable information. I will be presenting very recent work approaching the leader-follower problem...

With the widespread of data collection agent networks, distributed optimization and learning methods become preferable over centralized solutions. Typically, distributed machine learning problems are solved by having the network’s agents aim for a common (or consensus) model. In certain applications, however, each agent may be interested in meeting a personal goal which may differ from the consensus solution. This problem is referred to as (asynchronous)...

The Expectation-Maximization (EM) algorithm is one of the most popular methods used to solve the problem of distribution-based clustering in unsupervised learning. In this talk, we propose a dynamical systems perspective of the EM algorithm. More precisely, we can analyze the EM algorithm as a nonlinear state-space dynamical system. This algorithm belongs to a large class of iterative algorithms known as proximal point methods. In...

Feedzai is a scale up company with one mission: making banking and commerce safe. For that purpose, Feedzai develops methods for fraud prevention that should simultaneously be accurate, scalable and work within low latencies. In this talk will cover Feedzai’s research challenges, focusing on the use of neural networks for fraud prevention. For a long time, best solutions for fraud detection combined tree-based methods and handmade...

This work contributes an optimization framework in the context of structured interactions between an agent playing the role of a ‘provider’ and a human ‘receiver’. Examples of provider/receiver interactions of interest include ones between occupational therapist and patient, or teacher and student. We specifically consider tasks where the provider agent needs to plan a sequence of actions, where actions have associated costs and are organized...

Falls are very rare and extremely difficult to acquire in free living conditions. Due to this, most of prior work on fall detection has focused on simulated datasets acquired in scenarios that mimic the real-world context, however, the validation of systems trained with simulated falls remains unclear. This work presents a transfer learning approach for combining a dataset of simulated falls and non-falls, obtained from...

Structured representations are a powerful tool in machine learning, and in particular in natural language processing: The discrete, compositional nature of words and sentences leads to natural combinatorial representations such as trees, sequences, segments, or alignments, among others. At the same time, deep, hierarchical neural networks with latent representations are increasingly widely and successfully applied to language tasks. Deep networks conventionally perform smooth, soft computations...

Despite the recent success of state-of-the-art deep learning algorithms in object recognition, when these are deployed as-is on a mobile service robot, we observed that they failed to recognize many objects in real human environments. In this paper, we introduce a learning algorithm in which robots address this flaw by asking humans for help, also known as a symbiotic autonomy approach. In particular, we bootstrap...

Artificial Intelligence, and its diverse subfields, including machine learning, has been the subject of intense study for more than half a century. Recent advances in machine learning, jointly known as deep learning, have partially closed the gap that exists between the abilities of naturally intelligent systems (i.e., brains) and artificially intelligent ones in problems related with perception. Additionally, some problems that have been deemed very...