We propose a software framework based on the ideas of the Learning-Compression (LC) algorithm, that allows a user to compress a neural network or other machine learning model using different compression schemes with minimal effort. Currently, the supported compressions include pruning, quantization, low-rank methods (including automatically learning the layer ranks), and combinations of those, and the user can choose different compression types for different parts of a neural network. The LC algorithm alternates two types of steps until convergence: a learning (L) step, which trains a model on a dataset (using an algorithm such as SGD); and a compression (C) step, which compresses the model parameters (using a compression scheme such as low-rank or quantization). This decoupling of the "machine learning" aspect from the "signal compression" aspect means that changing the model or the compression type amounts to calling the corresponding subroutine in the L or C step, respectively. The library fully supports this by design, which makes it flexible and extensible. This does not come at the expense of performance: the runtime needed to compress a model is comparable to that of training the model in the first place; and the compressed model is competitive in terms of prediction accuracy and compression ratio with other algorithms (which are often specialized for specific models or compression schemes). The library is written in Python and PyTorch and available in Github.

Active learning is a framework in which the learning machine can select the samples to be used for training. This technique is promising, particularly when the cost of data acquisition and labeling is high. In active learning, determining the timing at which learning should be stopped is a critical issue. In this study, we propose a criterion for automatically stopping active learning. The proposed stopping criterion is based on the difference in the expected generalization errors and hypothesis testing. We derive a novel upper bound for the difference in expected generalization errors before and after obtaining a new training datum based on PAC-Bayesian theory. Unlike ordinary PAC-Bayesian bounds, though, the proposed bound is deterministic; hence, there is no uncontrollable trade-off between the confidence and tightness of the inequality. We combine the upper bound with a statistical test to derive a stopping criterion for active learning. We demonstrate the effectiveness of the proposed method via experiments with both artificial and real datasets.

The training dynamics of hidden layers in deep learning are poorly understood in theory. Recently, the Information Plane (IP) was proposed to analyze them, which is based on the information-theoretic concept of mutual information (MI). The Information Bottleneck (IB) theory predicts that layers maximize relevant information and compress irrelevant information. Due to the limitations in MI estimation from samples, there is an ongoing debate about the properties of the IP for the supervised learning case. In this work, we derive a theoretical convergence for the IP of autoencoders. The theory predicts that ideal autoencoders with a large bottleneck layer size do not compress input information, whereas a small size causes compression only in the encoder layers. For the experiments, we use a Gram-matrix based MI estimator recently proposed in the literature. We propose a new rule to adjust its parameters that compensates scale and dimensionality effects. Using our proposed rule, we obtain experimental IPs closer to the theory. Our theoretical IP for autoencoders could be used as a benchmark to validate new methods to estimate MI in neural networks. In this way, experimental limitations could be recognized and corrected, helping with the ongoing debate on the supervised learning case.

Currently, the divergence in distributions of design and operational data, and large computational complexity are limiting factors in the adoption of CNNs in real-world applications. For instance, person re-identification systems typically rely on a distributed set of cameras, where each camera has different capture conditions. This can translate to a considerable shift between source (e.g. lab setting) and target (e.g. operational camera) domains. Given the cost of annotating image data captured for fine-tuning in each target domain, unsupervised domain adaptation (UDA) has become a popular approach to adapt CNNs. Moreover, state-of-the-art deep learning models that provide a high level of accuracy often rely on architectures that are too complex for real-time applications. Although several compression and UDA approaches have recently been proposed to overcome these limitations, they do not allow optimizing a CNN to simultaneously address both. In this paper, we propose an unexplored direction -- the joint optimization of CNNs to provide a compressed model that is adapted to perform well for a given target domain. In particular, the proposed approach performs unsupervised knowledge distillation (KD) from a complex teacher model to a compact student model, by leveraging both source and target data. It also improves upon existing UDA techniques by progressively teaching the student about domain-invariant features, instead of directly adapting a compact model on target domain data. Our method is compared against state-of-the-art compression and UDA techniques, using two popular classification datasets for UDA -- Office31 and ImageClef-DA. In both datasets, results indicate that our method can achieve the highest level of accuracy while requiring a comparable or lower time complexity.

Meta-materials are an important emerging class of engineered materials in which complex macroscopic behaviour--whether electromagnetic, thermal, or mechanical--arises from modular substructure. Simulation and optimization of these materials are computationally challenging, as rich substructures necessitate high-fidelity finite element meshes to solve the governing PDEs. To address this, we leverage parametric modular structure to learn component-level surrogates, enabling cheaper high-fidelity simulation. We use a neural network to model the stored potential energy in a component given boundary conditions. This yields a structured prediction task: macroscopic behavior is determined by the minimizer of the system's total potential energy, which can be approximated by composing these surrogate models. Composable energy surrogates thus permit simulation in the reduced basis of component boundaries. Costly ground-truth simulation of the full structure is avoided, as training data are generated by performing finite element analysis with individual components. Using dataset aggregation to choose training boundary conditions allows us to learn energy surrogates which produce accurate macroscopic behavior when composed, accelerating simulation of parametric meta-materials.

This paper proposes a novel scalable community-based neural framework for graph learning. The framework learns the graph topology through the task of community detection and link prediction by optimizing with our proposed joint SBM loss function, which results from a non-trivial adaptation of the likelihood function of the classic Stochastic Block Model (SBM). Compared with SBM, our framework is flexible, naturally allows soft labels and digestion of complex node attributes. The main goal is efficient valuation of complex graph data, therefore our design carefully aims at accommodating large data, and ensures there is a single forward pass for efficient evaluation. For large graph, it remains an open problem of how to efficiently leverage its underlying structure for various graph learning tasks. Previously it can be heavy work. With our community-based framework, this becomes less difficult and allows the task models to basically plug-in-and-play and perform joint training. We currently look into two particular applications, the graph alignment and the anomalous correlation detection, and discuss how to make use of our framework to tackle both problems. Extensive experiments are conducted to demonstrate the effectiveness of our approach. We also contributed tweaks of classic techniques which we find helpful for performance and scalability. For example, 1) the GAT+, an improved design of GAT (Graph Attention Network), the scaled-cosine similarity, and a unified implementation of the convolution/attention based and the random-walk based neural graph models.

One way to analyse the behaviour of machine learning models is through local explanations that highlight input features that maximally influence model predictions. Sensitivity analysis, which involves analysing the effect of input perturbations on model predictions, is one of the methods to generate local explanations. Meaningful input perturbations are essential for generating reliable explanations, but there exists limited work on what such perturbations are and how to perform them. This work investigates these questions in the context of machine listening models that analyse audio. Specifically, we use a state-of-the-art deep singing voice detection (SVD) model to analyse whether explanations from SoundLIME (a local explanation method) are sensitive to how the method perturbs model inputs. The results demonstrate that SoundLIME explanations are sensitive to the content in the occluded input regions. We further propose and demonstrate a novel method for quantitatively identifying suitable content type(s) for reliably occluding inputs of machine listening models. The results for the SVD model suggest that the average magnitude of input mel-spectrogram bins is the most suitable content type for temporal explanations.

The estimation of patient-specific tissue properties in the form of model parameters is important for personalized physiological models. However, these tissue properties are spatially varying across the underlying anatomical model, presenting a significance challenge of high-dimensional (HD) optimization at the presence of limited measurement data. A common solution to reduce the dimension of the parameter space is to explicitly partition the anatomical mesh, either into a fixed small number of segments or a multi-scale hierarchy. This anatomy-based reduction of parameter space presents a fundamental bottleneck to parameter estimation, resulting in solutions that are either too low in resolution to reflect tissue heterogeneity, or too high in dimension to be reliably estimated within feasible computation. In this paper, we present a novel concept that embeds a generative variational auto-encoder (VAE) into the objective function of Bayesian optimization, providing an implicit low-dimensional (LD) search space that represents the generative code of the HD spatially-varying tissue properties. In addition, the VAE-encoded knowledge about the generative code is further used to guide the exploration of the search space. The presented method is applied to estimating tissue excitability in a cardiac electrophysiological model. Synthetic and real-data experiments demonstrate its ability to improve the accuracy of parameter estimation with more than 10x gain in efficiency.

Convolutional neural networks (CNNs) are widely used for image recognition and text analysis, and have been suggested for application on one-dimensional data as a way to reduce the need for pre-processing steps. Pre-processing is an integral part of multivariate analysis, but determination of the optimal pre-processing methods can be time-consuming due to the large number of available methods. In this work, the performance of a CNN was investigated for classification and regression analysis of spectral data. The CNN was compared with various other chemometric methods, including support vector machines (SVMs) for classification and partial least squares regression (PLSR) for regression analysis. The comparisons were made both on raw data, and on data that had gone through pre-processing and/or feature selection methods. The models were used on spectral data acquired with methods based on near-infrared, mid-infrared, and Raman spectroscopy. For the classification datasets the models were evaluated based on the percentage of correctly classified observations, while for regression analysis the models were assessed based on the coefficient of determination (R$^2$). Our results show that CNNs can outperform standard chemometric methods, especially for classification tasks where no pre-processing is used. However, both CNN and the standard chemometric methods see improved performance when proper pre-processing and feature selection methods are used. These results demonstrate some of the capabilities and limitations of CNNs used on one-dimensional data.

Recent developments in the filed of Deep Learning have demonstrated that Deep Neural Networks(DNNs) are vulnerable to adversarial examples. Specifically, in image classification, an adversarial example can fool the well trained deep neural networks by adding barely imperceptible perturbations to clean images. Adversarial Training, one of the most direct and effective methods, minimizes the losses of perturbed-data to learn robust deep networks against adversarial attacks. It has been proven that using the fast gradient sign method (FGSM) can achieve Fast Adversarial Training. However, FGSM-based adversarial training may finally obtain a failed model because of overfitting to FGSM samples. In this paper, we proposed the Diversified Initialized Perturbations Adversarial Training (DIP-FAT) which involves seeking the initialization of the perturbation via enlarging the output distances of the target model in a random directions. Due to the diversity of random directions, the embedded fast adversarial training using FGSM increases the information from the adversary and reduces the possibility of overfitting. In addition to preventing overfitting, the extensive results show that our proposed DIP-FAT technique can also improve the accuracy of the clean data. The biggest advantage of DIP-FAT method: achieving the best banlance among clean-data, perturbed-data and efficiency.