Multivariate time series forecasting for freeway networks

This dissertation introduces traffic forecasting methods for different network configurations and data availability.Chapter 2 focuses on single freeway cases.Although its topology is simple, the non-linearity of traffic features makes this prediction still a challenging task.We propose the dynamic linear model (DLM) to approximate the non-linear traffic features. Unlike a static linear regression model, the DLM assumes that its parameters change over time.We design the DLM with time-dependent model parameters to describe the spatiotemporal characteristics of time-series traffic data. Based on our DLM and its model parameters analytically trained using historical data, we suggest the optimal linear predictor in the minimum mean square error (MMSE) sense.We compare our prediction accuracy by estimating expected travel time based on the traffic prediction for freeways in California (I210-E and I5-S) under highly congested traffic conditions with other baselines. We show significant improvements in accuracy, especially for short-term prediction.Chapter 3 aims to generalize the DLM to extensive freeway networks with more complex topologies.Most resources would be consumed to estimate unnecessary spatiotemporal correlations if the DLM was directly used for a large-scale network.Defining features on graphs relaxes such issues by cutting unnecessary connections in advance based on predefined topology information.Exploiting the graph signal processing, we represent traffic dynamics over freeway networks using multiple graph heat diffusion kernels and integrate the kernels into DLM with Bayes' rule. We optimize the model parameters using Bayesian inference to minimize the prediction errors.The proposed model demonstrates prediction accuracy comparable to state-of-the-art deep neural networks with lower computational effort. It notably achieves excellent performance for long-term prediction through the inheritance of periodicity modeling in DLM.Chapter 4 proposes a deep neural network model to predict traffic features on large-scale freeway networks.These days, deep learning methods have heavily tackled traffic forecasting problems of freeway networks because they are outstanding at learning highly complex correlations between variables both in time and space, which the linear models might be limited to.Adopting a graph convolutional network (GCN) becomes a standard to extract spatial correlations; therefore, most works have achieved great prediction accuracy by implanting it into their architecture.However, the conventional GCN has the drawback that receptive field size should be small, i.e., barely refers to traffic features of remote sensors, resulting in inaccurate long-term prediction.We suggest a forecasting model called two-level resolution deep neural network (TwoResNet) that overcomes the limitation.It consists of two resolution blocks:The low-resolution block predicts traffic on a macroscopic scale, such as regional traffic changes.On the other hand, the high-resolution block predicts traffic on a microscopic scale by using GCN to extract spatial correlations, referring to the regional changes produced by the low-resolution block.This process allows the GCN to refer to the traffic features from remote sensors.As a result, TwoResNet achieves competitive prediction accuracy compared to state-of-the-art methods, especially showing excellent performance for long-term predictions.

Theory of representation learning in cortical neural networks

Carlos Stein Naves de Brito

Our brain continuously self-organizes to construct and maintain an internal representation of the world based on the information arriving through sensory stimuli. Remarkably, cortical areas related to different sensory modalities appear to share the same functional unit, the neuron, and develop through the same learning mechanism, synaptic plasticity. It motivates the conjecture of a unifying theory to explain cortical representational learning across sensory modalities. In this thesis we present theories and computational models of learning and optimization in neural networks, postulating functional properties of synaptic plasticity that support the apparent universal learning capacity of cortical networks. In the past decades, a variety of theories and models have been proposed to describe receptive field formation in sensory areas. They include normative models such as sparse coding, and bottom-up models such as spike-timing dependent plasticity. We bring together candidate explanations by demonstrating that in fact a single principle is sufficient to explain receptive field development. First, we show that many representative models of sensory development are in fact implementing variations of a common principle: nonlinear Hebbian learning. Second, we reveal that nonlinear Hebbian learning is sufficient for receptive field formation through sensory inputs. A surprising result is that our findings are independent of specific details, and allow for robust predictions of the learned receptive fields. Thus nonlinear Hebbian learning and natural statistics can account for many aspects of receptive field formation across models and sensory modalities. The Hebbian learning theory substantiates that synaptic plasticity can be interpreted as an optimization procedure, implementing stochastic gradient descent. In stochastic gradient descent inputs arrive sequentially, as in sensory streams. However, individual data samples have very little information about the correct learning signal, and it becomes a fundamental problem to know how many samples are required for reliable synaptic changes. Through estimation theory, we develop a novel adaptive learning rate model, that adapts the magnitude of synaptic changes based on the statistics of the learning signal, enabling an optimal use of data samples. Our model has a simple implementation and demonstrates improved learning speed, making this a promising candidate for large artificial neural network applications. The model also makes predictions on how cortical plasticity may modulate synaptic plasticity for optimal learning. The optimal sampling size for reliable learning allows us to estimate optimal learning times for a given model. We apply this theory to derive analytical bounds on times for the optimization of synaptic connections. First, we show this optimization problem to have exponentially many saddle-nodes, which lead to small gradients and slow learning. Second, we show that the number of input synapses to a neuron modulates the magnitude of the initial gradient, determining the duration of learning. Our final result reveals that the learning duration increases supra-linearly with the number of synapses, suggesting an effective limit on synaptic connections and receptive field sizes in developing neural networks.

EPFL2016

Evaluating and Interpreting Deep Convolutional Neural Networks via Non-negative Matrix Factorization

Edo Collins

With ever greater computational resources and more accessible software, deep neural networks have become ubiquitous across industry and academia. Their remarkable ability to generalize to new samples defies the conventional view, which holds that complex, over-parameterized networks would be prone to overfitting. This apparent discrepancy is exacerbated by our inability to inspect and interpret the high-dimensional, non-linear, latent representations they learn, which has led many to refer to neural networks as black-boxes''. The Law of Parsimony states that simpler solutions are more likely to be correct than complex ones''. Since they perform quite well in practice, a natural question to ask, then, is in what way are neural networks simple? We propose that compression is the answer. Since good generalization requires invariance to irrelevant variations in the input, it is necessary for a network to discard this irrelevant information. As a result, semantically similar samples are mapped to similar representations in neural network deep feature space, where they form simple, low-dimensional structures. Conversely, a network that overfits relies on memorizing individual samples. Such a network cannot discard information as easily. In this thesis we characterize the difference between such networks using the non-negative rank of activation matrices. Relying on the non-negativity of rectified-linear units, the non-negative rank is the smallest number that admits an exact non-negative matrix factorization. We derive an upper bound on the amount of memorization in terms of the non-negative rank, and show it is a natural complexity measure for rectified-linear units. With a focus on deep convolutional neural networks trained to perform object recognition, we show that the two non-negative factors derived from deep network layers decompose the information held therein in an interpretable way. The first of these factors provides heatmaps which highlight similarly encoded regions within an input image or image set. We find that these networks learn to detect semantic parts and form a hierarchy, such that parts are further broken down into sub-parts. We quantitatively evaluate the semantic quality of these heatmaps by using them to perform semantic co-segmentation and co-localization. In spite of the convolutional network we use being trained solely with image-level labels, we achieve results comparable or better than domain-specific state-of-the-art methods for these tasks. The second non-negative factor provides a bag-of-concepts representation for an image or image set. We use this representation to derive global image descriptors for images in a large collection. With these descriptors in hand, we perform two variations content-based image retrieval, i.e. reverse image search. Using information from one of the non-negative matrix factors we obtain descriptors which are suitable for finding semantically related images, i.e., belonging to the same semantic category as the query image. Combining information from both non-negative factors, however, yields descriptors that are suitable for finding other images of the specific instance depicted in the query image, where we again achieve state-of-the-art performance.

EPFL2019

Biologically plausible unsupervised learning in shallow and deep neural networks

Bernd Albert Illing

The way our brain learns to disentangle complex signals into unambiguous concepts is fascinating but remains largely unknown. There is evidence, however, that hierarchical neural representations play a key role in the cortex. This thesis investigates biologically plausible models of unsupervised learning of hierarchical representations as found in the brain and modern computer vision models. We use computational modeling to address three main questions at the intersection of artificial intelligence (AI) and computational neuroscience.The first question is: What are useful neural representations and when are deep hierarchical representations needed? We approach this point with a systematic study of biologically plausible unsupervised feature learning in a shallow 2-layer networks on digit (MNIST) and object (CIFAR10) classification. Surprisingly, random features support high performance, especially for large hidden layers. When combined with localized receptive fields, random feature networks approach the performance of supervised backpropagation on MNIST, but not on CIFAR10. We suggest that future models of biologically plausible learning should outperform such random feature benchmarks on MNIST, or that such models should be evaluated in different ways.The second question is: How can hierarchical representations be learned with mechanisms supported by neuroscientific evidence? We cover this question by proposing a unifying Hebbian model, inspired by common models of V1 simple and complex cells based on unsupervised sparse coding and temporal invariance learning. In shallow 2-layer networks, our model reproduces learning of simple and complex cell receptive fields, as found in V1. In deeper networks, we stack multiple layers of Hebbian learning but find that it does not yield hierarchical representations of increasing usefulness. From this, we hypothesise that standard Hebbian rules are too constrained to build increasingly useful representations, as observed in higher areas of the visual cortex or deep artificial neural networks.The third question is: Can AI inspire learning models that build deep representations and are still biologically plausible? We address this question by proposing a learning rule that takes inspiration from neuroscience and recent advances in self-supervised deep learning. The proposed rule is Hebbian, i.e. only depends on pre- and post-synaptic neuronal activity, but includes additional local factors, namely predictive dendritic input and widely broadcasted modulation factors. Algorithmically, this rule applies self-supervised contrastive predictive learning to a causal, biological setting using saccades. We find that networks trained with this generalised Hebbian rule build deep hierarchical representations of images, speech and video.We see our modeling as a potential starting point for both, new hypotheses, that can be tested experimentally, and novel AI models that could benefit from added biological realism.

EPFL2021