Recent work has shown the tremendous vulnerability to adversarial samples that are nearly indistinguishable from benign data but are improperly classified by the deep learning model. Some of the latest findings suggest the existence of adversarial attacks may be an inherent weakness of these models as a direct result of its sensitivity to well-generalizing features in high dimensional data. We hypothesize that data transformations can influence this vulnerability since a change in the data manifold directly determines the adversary's ability to create these adversarial samples. To approach this problem, we study the effect of dimensionality reduction through the lens of adversarial robustness. This study raises awareness of the positive and negative impacts of five commonly used data transformation techniques on adversarial robustness. The evaluation shows how these techniques contribute to an overall increased vulnerability where accuracy is only improved when the dimensionality reduction technique approaches the data's optimal intrinsic dimension. The conclusions drawn from this work contribute to understanding and creating more resistant learning models.

For instance, Tucker (Tucker, Tensor decomposition methods have recently 1966) and canonical polyadic (CP) (Caroll & Chang, 1970) proven to be efficient for compressing and accelerating tensor decompositions are widely known for compressing neural networks. However, the problem and accelerating convolutional networks (Lebedev of optimal decomposition structure determination et al., 2015; Kim et al., 2016; Kossaifi et al., 2019), and is still not well studied while being quite important. Tensor Train (TT) (Oseledets, 2011) decomposition has Specifically, decomposition ranks present been successfully applied for compressing fully-connected the crucial parameter controlling the compressionaccuracy (FC) (Novikov et al., 2015), convolutional (Garipov et al., tradeoff. In this paper, we introduce 2016), recurrent (Yang et al., 2017; Yu et al., 2017), embedding MARS -- a new efficient method for the automatic (Khrulkov et al., 2019) layers.

Standard causal discovery methods must fit a new model whenever they encounter samples from a new underlying causal graph. However, these samples often share relevant information - for instance, the dynamics describing the effects of causal relations - which is lost when following this approach. We propose Amortized Causal Discovery, a novel framework that leverages such shared dynamics to learn to infer causal relations from time-series data. This enables us to train a single, amortized model that infers causal relations across samples with different underlying causal graphs, and thus makes use of the information that is shared. We demonstrate experimentally that this approach, implemented as a variational model, leads to significant improvements in causal discovery performance, and show how it can be extended to perform well under hidden confounding.

We provide an empirical study of the stability of recurrent neural networks trained to recognize regular languages. When a small amount of noise is introduced into the activation function, the neurons in the recurrent layer tend to saturate in order to compensate the variability. In this saturated regime, analysis of the network activation shows a set of clusters that resemble discrete states in a finite state machine. We show that transitions between these states in response to input symbols are deterministic and stable. The networks display a stable behavior for arbitrarily long strings, and when random perturbations are applied to any of the states, they are able to recover and their evolution converges to the original clusters. This observation reinforces the interpretation of the networks as finite automata, with neurons or groups of neurons coding specific and meaningful input patterns.

Sensory learning in the mammalian cortex has long been hypothesized to involve the objective of variational inference (VI). Likely the most well-known algorithm for cortical VI is the Wake-Sleep algorithm (Hinton et al. 1995). However Wake-Sleep problematically assumes that neural activities are independent given lower-layers during generation. Here, we construct a VI system that is both compatible with neurobiology and avoids this assumption. The core of the system is a wake-sleep discriminator that classifies network states as inferred or self-generated. Inference connections learn by opposing this discriminator. This adversarial dynamic solves a core problem within VI, which is to match the distribution of stimulus-evoked (inference) activity to that of self-generated activity. Meanwhile, generative connections learn to predict lower-level activity as in standard VI. We implement this algorithm and show that it can successfully train the approximate inference network for generative models. Our proposed algorithm makes several biological predictions that can be tested. Most importantly, it predicts a teaching signal that is remarkably similar to known properties of the cholinergic system.

We investigate a strategy for improving the computational efficiency of contrastive learning of visual representations by leveraging a small amount of supervised information during pre-training. We propose a semi-supervised loss, SuNCEt, based on noise-contrastive estimation, that aims to distinguish examples of different classes in addition to the self-supervised instance-wise pretext tasks. We find that SuNCEt can be used to match the semi-supervised learning accuracy of previous contrastive approaches with significantly less computational effort. Our main insight is that leveraging even a small amount of labeled data during pre-training, and not only during fine-tuning, provides an important signal that can significantly accelerate contrastive learning of visual representations.

Understanding patients' journeys in healthcare system is a fundamental prepositive task for a broad range of AI-based healthcare applications. This task aims to learn an informative representation that can comprehensively encode hidden dependencies among medical events and its inner entities, and then the use of encoding outputs can greatly benefit the downstream application-driven tasks. A patient journey is a sequence of electronic health records (EHRs) over time that is organized at multiple levels: patient, visits and medical codes. The key challenge of patient journey understanding is to design an effective encoding mechanism which can properly tackle the aforementioned multi-level structured patient journey data with temporal sequential visits and a set of medical codes. This paper proposes a novel self-attention mechanism that can simultaneously capture the contextual and temporal relationships hidden in patient journeys. A multi-level self-attention network (MusaNet) is specifically designed to learn the representations of patient journeys that is used to be a long sequence of activities. The MusaNet is trained in end-to-end manner using the training data derived from EHRs. We evaluated the efficacy of our method on two medical application tasks with real-world benchmark datasets. The results have demonstrated the proposed MusaNet produces higher-quality representations than state-of-the-art baseline methods. The source code is available in https://github.com/xueping/MusaNet.

Recovery after stroke is often incomplete, but rehabilitation training may potentiate recovery by engaging endogenous neuroplasticity. In preclinical models of stroke, high doses of rehabilitation training are required to restore functional movement to the affected limbs of animals. In humans, however, the necessary dose of training to potentiate recovery is not known. This ignorance stems from the lack of objective, pragmatic approaches for measuring training doses in rehabilitation activities. Here, to develop a measurement approach, we took the critical first step of automatically identifying functional primitives, the basic building block of activities. Forty-eight individuals with chronic stroke performed a variety of rehabilitation activities while wearing inertial measurement units (IMUs) to capture upper body motion. Primitives were identified by human labelers, who labeled and segmented the associated IMU data. We performed automatic classification of these primitives using machine learning. We designed a convolutional neural network model that outperformed existing methods. The model includes an initial module to compute separate embeddings of different physical quantities in the sensor data. In addition, it replaces batch normalization (which performs normalization based on statistics computed from the training data) with instance normalization (which uses statistics computed from the test data). This increases robustness to possible distributional shifts when applying the method to new patients. With this approach, we attained an average classification accuracy of 70%. Thus, using a combination of IMU-based motion capture and deep learning, we were able to identify primitives automatically. This approach builds towards objectively-measured rehabilitation training, enabling the identification and counting of functional primitives that accrues to a training dose.

Zero-shot learning relies on semantic class representations such as attributes or pretrained embeddings to predict classes without any labeled examples. We propose to learn class representations from common sense knowledge graphs. Common sense knowledge graphs are an untapped source of explicit high-level knowledge that requires little human effort to apply to a range of tasks. To capture the knowledge in the graph, we introduce ZSL-KG, a framework based on graph neural networks with non-linear aggregators to generate class representations. Whereas most prior work on graph neural networks uses linear functions to aggregate information from neighboring nodes, we find that non-linear aggregators such as LSTMs or transformers lead to significant improvements on zero-shot tasks. On two natural language tasks across three datasets, ZSL-KG shows an average improvement of 9.2 points of accuracy versus state-of-the-art methods. In addition, on an object classification task, ZSL-KG shows a 2.2 accuracy point improvement versus the best methods that do not require hand-engineered class representations. Finally, we find that ZSL-KG outperforms the best performing graph neural networks with linear aggregators by an average of 3.8 points of accuracy across these four datasets.

Deep Neural Networks (DNNs) are often criticized for being susceptible to adversarial attacks. Most successful defense strategies adopt adversarial training or random input transformations that typically require retraining or fine-tuning the model to achieve reasonable performance. In this work, our investigations of intermediate representations of a pre-trained DNN lead to an interesting discovery pointing to intrinsic robustness to adversarial attacks. We find that we can learn a generative classifier by statistically characterizing the neural response of an intermediate layer to clean training samples. The predictions of multiple such intermediate-layer based classifiers, when aggregated, show unexpected robustness to adversarial attacks. Specifically, we devise an ensemble of these generative classifiers that rank-aggregates their predictions via a Borda count-based consensus. Our proposed approach uses a subset of the clean training data and a pre-trained model, and yet is agnostic to network architectures or the adversarial attack generation method. We show extensive experiments to establish that our defense strategy achieves state-of-the-art performance on the ImageNet validation set.