Catastrophic Forgetting in Deep Graph Networks: an Introductory Benchmark for Graph Classification

In this work, we study the phenomenon of catastrophic forgetting in the graph representation learning scenario. The primary objective of the analysis is to understand whether classical continual learning techniques for flat and sequential data have a tangible impact on performances when applied to graph data. To do so, we experiment with a structure-agnostic model and a deep graph network in a robust and controlled environment on three different datasets. The benchmark is complemented by an investigation on the effect of structure-preserving regularization techniques on catastrophic forgetting. We find that replay is the most effective strategy in so far, which also benefits the most from the use of regularization. Our findings suggest interesting future research at the intersection of the continual and graph representation learning fields. Finally, we provide researchers with a flexible software framework to reproduce our results and carry out further experiments.

Continual Learning for Recurrent Neural Networks: a Review and Empirical Evaluation

Learning continuously during all model lifetime is fundamental to deploy machine learning solutions robust to drifts in the data distribution. Advances in Continual Learning (CL) with recurrent neural networks could pave the way to a large number of applications where incoming data is non stationary, like natural language processing and robotics. However, the existing body of work on the topic is still fragmented, with approaches which are application-specific and whose assessment is based on heterogeneous learning protocols and datasets. In this paper, we organize the literature on CL for sequential data processing by providing a categorization of the contributions and a review of the benchmarks. We propose two new benchmarks for CL with sequential data based on existing datasets, whose characteristics resemble real-world applications. We also provide a broad empirical evaluation of CL and Recurrent Neural Networks in class-incremental scenario, by testing their ability to mitigate forgetting with a number of different strategies which are not specific to sequential data processing. Our results highlight the key role played by the sequence length and the importance of a clear specification of the CL scenario.

Graph Mixture Density Networks

We introduce the Graph Mixture Density Network, a new family of machine learning models that can fit multimodal output distributions conditioned on arbitrary input graphs. By combining ideas from mixture models and graph representation learning, we address a broad class of challenging regression problems that rely on structured data. Our main contribution is the design and evaluation of our method on large stochastic epidemic simulations conditioned on random graphs. We show that there is a significant improvement in the likelihood of an epidemic outcome when taking into account both multimodality and structure. In addition, we investigate how to \textit{implicitly} retain structural information in node representations by computing the distance between distributions of adjacent nodes, and the technique is tested on two structure reconstruction tasks with very good accuracy. Graph Mixture Density Networks open appealing research opportunities in the study of structure-dependent phenomena that exhibit non-trivial conditional output distributions.

Explaining Deep Graph Networks with Molecular Counterfactuals

We present a novel approach to tackle explainability of deep graph networks in the context of molecule property prediction tasks, named MEG (Molecular Explanation Generator). We generate informative counterfactual explanations for a specific prediction under the form of (valid) compounds with high structural similarity and different predicted properties. We discuss preliminary results showing how the model can convey non-ML experts with key insights into the learning model focus in the neighborhood of a molecule.

Short-Term Memory Optimization in Recurrent Neural Networks by Autoencoder-based Initialization

Training RNNs to learn long-term dependencies is difficult due to vanishing gradients. We explore an alternative solution based on explicit memorization using linear autoencoders for sequences, which allows to maximize the short-term memory and that can be solved with a closed-form solution without backpropagation. We introduce an initialization schema that pretrains the weights of a recurrent neural network to approximate the linear autoencoder of the input sequences and we show how such pretraining can better support solving hard classification tasks with long sequences. We test our approach on sequential and permuted MNIST. We show that the proposed approach achieves a much lower reconstruction error for long sequences and a better gradient propagation during the finetuning phase.

Learning from Non-Binary Constituency Trees via Tensor Decomposition

Processing sentence constituency trees in binarised form is a common and popular approach in literature. However, constituency trees are non-binary by nature. The binarisation procedure changes deeply the structure, furthering constituents that instead are close. In this work, we introduce a new approach to deal with non-binary constituency trees which leverages tensor-based models. In particular, we show how a powerful composition function based on the canonical tensor decomposition can exploit such a rich structure. A key point of our approach is the weight sharing constraint imposed on the factor matrices, which allows limiting the number of model parameters. Finally, we introduce a Tree-LSTM model which takes advantage of this composition function and we experimentally assess its performance on different NLP tasks.

Generative Tomography Reconstruction

We propose an end-to-end differentiable architecture for tomography reconstruction that directly maps a noisy sinogram into a denoised reconstruction. Compared to existing approaches our end-to-end architecture produces more accurate reconstructions while using less parameters and time. We also propose a generative model that, given a noisy sinogram, can sample realistic reconstructions. This generative model can be used as prior inside an iterative process that, by taking into consideration the physical model, can reduce artifacts and errors in the reconstructions.

FADER: Fast Adversarial Example Rejection

Deep neural networks are vulnerable to adversarial examples, i.e., carefully-crafted inputs that mislead classification at test time. Recent defenses have been shown to improve adversarial robustness by detecting anomalous deviations from legitimate training samples at different layer representations - a behavior normally exhibited by adversarial attacks. Despite technical differences, all aforementioned methods share a common backbone structure that we formalize and highlight in this contribution, as it can help in identifying promising research directions and drawbacks of existing methods. The first main contribution of this work is the review of these detection methods in the form of a unifying framework designed to accommodate both existing defenses and newer ones to come. In terms of drawbacks, the overmentioned defenses require comparing input samples against an oversized number of reference prototypes, possibly at different representation layers, dramatically worsening the test-time efficiency. Besides, such defenses are typically based on ensembling classifiers with heuristic methods, rather than optimizing the whole architecture in an end-to-end manner to better perform detection. As a second main contribution of this work, we introduce FADER, a novel technique for speeding up detection-based methods. FADER overcome the issues above by employing RBF networks as detectors: by fixing the number of required prototypes, the runtime complexity of adversarial examples detectors can be controlled. Our experiments outline up to 73x prototypes reduction compared to analyzed detectors for MNIST dataset and up to 50x for CIFAR10 dataset respectively, without sacrificing classification accuracy on both clean and adversarial data.

Perplexity-free Parametric t-SNE

The t-distributed Stochastic Neighbor Embedding (t-SNE) algorithm is a ubiquitously employed dimensionality reduction (DR) method. Its non-parametric nature and impressive efficacy motivated its parametric extension. It is however bounded to a user-defined perplexity parameter, restricting its DR quality compared to recently developed multi-scale perplexity-free approaches. This paper hence proposes a multi-scale parametric t-SNE scheme, relieved from the perplexity tuning and with a deep neural network implementing the mapping. It produces reliable embeddings with out-of-sample extensions, competitive with the best perplexity adjustments in terms of neighborhood preservation on multiple data sets.

ROS-Neuro Integration of Deep Convolutional Autoencoders for EEG Signal Compression in Real-time BCIs

Typical EEG-based BCI applications require the computation of complex functions over the noisy EEG channels to be carried out in an efficient way. Deep learning algorithms are capable of learning flexible nonlinear functions directly from data, and their constant processing latency is perfect for their deployment into online BCI systems. However, it is crucial for the jitter of the processing system to be as low as possible, in order to avoid unpredictable behaviour that can ruin the system's overall usability. In this paper, we present a novel encoding method, based on on deep convolutional autoencoders, that is able to perform efficient compression of the raw EEG inputs. We deploy our model in a ROS-Neuro node, thus making it suitable for the integration in ROS-based BCI and robotic systems in real world scenarios. The experimental results show that our system is capable to generate meaningful compressed encoding preserving to original information contained in the raw input. They also show that the ROS-Neuro node is able to produce such encodings at a steady rate, with minimal jitter. We believe that our system can represent an important step towards the development of an effective BCI processing pipeline fully standardized in ROS-Neuro framework.