We present a brief history of the field of interpretable machine learning (IML), give an overview of state-of-the-art interpretation methods, and discuss challenges. Research in IML has boomed in recent years. As young as the field is, it has over 200 years old roots in regression modeling and rule-based machine learning, starting in the 1960s. Recently, many new IML methods have been proposed, many of them model-agnostic, but also interpretation techniques specific to deep learning and tree-based ensembles. IML methods either directly analyze model components, study sensitivity to input perturbations, or analyze local or global surrogate approximations of the ML model. The field approaches a state of readiness and stability, with many methods not only proposed in research, but also implemented in open-source software. But many important challenges remain for IML, such as dealing with dependent features, causal interpretation, and uncertainty estimation, which need to be resolved for its successful application to scientific problems. A further challenge is a missing rigorous definition of interpretability, which is accepted by the community. To address the challenges and advance the field, we urge to recall our roots of interpretable, data-driven modeling in statistics and (rule-based) ML, but also to consider other areas such as sensitivity analysis, causal inference, and the social sciences.

Graph neural network models have been extensively used to learn node representations for graph structured data in an end-to-end setting. These models often rely on localized first order approximations of spectral graph convolutions and hence are unable to capture higher-order relational information between nodes. Probabilistic Graphical Models form another class of models that provide rich flexibility in incorporating such relational information but are limited by inefficient approximate inference algorithms at higher order. In this paper, we propose to combine these approaches to learn better node and graph representations. First, we derive an efficient approximate sum-product loopy belief propagation inference algorithm for higher-order PGMs. We then embed the message passing updates into a neural network to provide the inductive bias of the inference algorithm in end-to-end learning. This gives us a model that is flexible enough to accommodate domain knowledge while maintaining the computational advantage. We further propose methods for constructing higher-order factors that are conditioned on node and edge features and share parameters wherever necessary. Our experimental evaluation shows that our model indeed captures higher-order information, substantially outperforming state-of-the-art $k$-order graph neural networks in molecular datasets.

Lossy gradient compression has become a practical tool to overcome the communication bottleneck in centrally coordinated distributed training of machine learning models. However, algorithms for decentralized training with compressed communication over arbitrary connected networks have been more complicated, requiring additional memory and hyperparameters. We introduce a simple algorithm that directly compresses the model differences between neighboring workers using low-rank linear compressors applied on model differences. Inspired by the PowerSGD algorithm for centralized deep learning, this algorithm uses power iteration steps to maximize the information transferred per bit. We prove that our method requires no additional hyperparameters, converges faster than prior methods, and is asymptotically independent of both the network and the compression. Out of the box, these compressors perform on par with state-of-the-art tuned compression algorithms in a series of deep learning benchmarks.

This paper proposes an encoder-decoder architecture grounded on Generative Adversarial Networks (GANs), that generates a sequence of activities and their timestamps in an end-to-end way. GANs work well with differentiable data such as images. However, a suffix is a sequence of categorical items. To this end, we use the Gumbel-Softmax distribution to get a differentiable continuous approximation. The training works by putting one neural network against the other in a two-player game (hence the "adversarial" nature), which leads to generating suffixes close to the ground truth. From the experimental evaluation it emerges that the approach is superior to the baselines in terms of the accuracy of the predicted suffixes and corresponding remaining times, despite using a naive feature encoding and only engineering features based on control flow and events completion time.

Despite achieving strong performance in the semi-supervised node classification task, graph neural networks (GNNs) are vulnerable to adversarial attacks, similar to other deep learning models. Existing research works either focus on developing robust GNN models or attack detection methods against attacks on graphs. However, little research attention is paid to the potential and practice of immunization to adversarial attacks on graphs. In this paper, we formulate the problem of graph adversarial immunization as a bilevel optimization problem, i.e., vaccinating an affordable fraction of node pairs, connected or unconnected, to improve the certifiable robustness of the graph against any admissible adversarial attack. We further propose an efficient algorithm, called AdvImmune, which optimizes meta-gradient in a discrete way to circumvent the computationally expensive combinatorial optimization when solving the adversarial immunization problem. Experiments are conducted on two citation networks and one social network. Experimental results demonstrate that the proposed AdvImmune immunization method remarkably improves the fraction of robust nodes by 12%, 42%, 65%, with an affordable immune budget of only 5% edges.

To improve the sample efficiency of policy-gradient based reinforcement learning algorithms, we propose implicit distributional actor-critic (IDAC) that consists of a distributional critic, built on two deep generator networks (DGNs), and a semi-implicit actor (SIA), powered by a flexible policy distribution. We adopt a distributional perspective on the discounted cumulative return and model it with a state-action-dependent implicit distribution, which is approximated by the DGNs that take state-action pairs and random noises as their input. Moreover, we use the SIA to provide a semi-implicit policy distribution, which mixes the policy parameters with a reparameterizable distribution that is not constrained by an analytic density function. In this way, the policy's marginal distribution is implicit, providing the potential to model complex properties such as covariance structure and skewness, but its parameter and entropy can still be estimated. We incorporate these features with an off-policy algorithm framework to solve problems with continuous action space and compare IDAC with state-of-the-art algorithms on representative OpenAI Gym environments. We observe that IDAC outperforms these baselines in most tasks.

Normalizing flows, autoregressive models, variational autoencoders (VAEs), and deep energy-based models are among competing likelihood-based frameworks for deep generative learning. Among them, VAEs have the advantage of fast and tractable sampling and easy-to-access encoding networks. However, they are currently outperformed by other models such as normalizing flows and autoregressive models. While the majority of the research in VAEs is focused on the statistical challenges, we explore the orthogonal direction of carefully designing neural architectures for hierarchical VAEs. We propose Nouveau VAE (NVAE), a deep hierarchical VAE built for image generation using depth-wise separable convolutions and batch normalization. NVAE is equipped with a residual parameterization of Normal distributions and its training is stabilized by spectral regularization. We show that NVAE achieves state-of-the-art results among non-autoregressive likelihood-based models on the MNIST, CIFAR-10, CelebA 64, and CelebA HQ datasets and it provides a strong baseline on FFHQ. For example, on CIFAR-10, NVAE pushes the state-of-the-art from 2.98 to 2.91 bits per dimension, and it produces high-quality images on CelebA HQ. To the best of our knowledge, NVAE is the first successful VAE applied to natural images as large as 256$\times$256 pixels. The source code is available at https://github.com/NVlabs/NVAE .

This paper presents and characterizes an Open Application Repository for Federated Learning (OARF), a benchmark suite for federated machine learning systems. Previously available benchmarks for federated learning have focused mainly on synthetic datasets and use a very limited number of applications. OARF includes different data partitioning methods (horizontal, vertical and hybrid) as well as emerging applications in image, text and structured data, which represent different scenarios in federated learning. Our characterization shows that the benchmark suite is diverse in data size, distribution, feature distribution and learning task complexity. We have developed reference implementations, and evaluated the important aspects of federated learning, including model accuracy, communication cost, differential privacy, secure multiparty computation and vertical federated learning.

Graph Neural Networks (GNNs) have demonstrated superior performance in learning graph representations for several subsequent downstream inference tasks. However, learning over graph data can raise privacy concerns when nodes represent people or human-related variables that involve personal information about individuals. Previous works have presented various techniques for privacy-preserving deep learning over non-relational data, such as image, audio, video, and text, but there is less work addressing the privacy issues involved in applying deep learning algorithms on graphs. As a result and for the first time, in this paper, we develop a privacy-preserving GNN learning algorithm with formal privacy guarantees based on Local Differential Privacy (LDP) to tackle the problem of node-level privacy, where graph nodes have potentially sensitive features that need to be kept private, but they could be beneficial for an untrusted server to learn richer node representations. Specifically, we propose an optimized LDP algorithm with an unbiased estimator, using which a central server can communicate with the graph nodes to privately collect their data and estimate the graph convolution layer of the GNN. To further reduce the effect of the injected noise, we propose a simple graph convolution layer based on the multi-hop aggregation of the nodes' features. Extensive experiments conducted over real-world datasets demonstrate the capability of our method in maintaining an appropriate privacy-accuracy trade-off for privacy-preserving node classification.

Recurrent neural networks (RNNs) with continuous-time hidden states are a natural fit for modeling irregularly-sampled time series. These models, however, face difficulties when the input data possess long-term dependencies. We prove that similar to standard RNNs, the underlying reason for this issue is the vanishing or exploding of the gradient during training. This phenomenon is expressed by the ordinary differential equation (ODE) representation of the hidden state, regardless of the ODE solver's choice. We provide a solution by designing a new algorithm based on the long short-term memory (LSTM) that separates its memory from its time-continuous state. This way, we encode a continuous-time dynamical flow within the RNN, allowing it to respond to inputs arriving at arbitrary time-lags while ensuring a constant error propagation through the memory path. We call these RNN models ODE-LSTMs. We experimentally show that ODE-LSTMs outperform advanced RNN-based counterparts on non-uniformly sampled data with long-term dependencies. All code and data is available at https://github.com/mlech26l/learning-long-term-irregular-ts.