Deep reinforcement learning (DRL) is a booming area of artificial intelligence. Many practical applications of DRL naturally involve more than one collaborative learners, making it important to study DRL in a multi-agent context. Previous research showed that effective learning in complex multi-agent systems demands for highly coordinated environment exploration among all the participating agents. Many researchers attempted to cope with this challenge through learning centralized value functions. However, the common strategy for every agent to learn their local policies directly often fail to nurture strong inter-agent collaboration and can be sample inefficient whenever agents alter their communication channels. To address these issues, we propose a new framework known as centralized training and exploration with decentralized execution via policy distillation. Guided by this framework and the maximum-entropy learning technique, we will first train agents' policies with shared global component to foster coordinated and effective learning. Locally executable policies will be derived subsequently from the trained global policies via policy distillation. Experiments show that our new framework and algorithm can achieve significantly better performance and higher sample efficiency than a cutting-edge baseline on several multi-agent DRL benchmarks.

Recent studies on the adversarial vulnerability of neural networks have shown that models trained with the objective of minimizing an upper bound on the worst-case loss over all possible adversarial perturbations improve robustness against adversarial attacks. Beside exploiting adversarial training framework, we show that by enforcing a Deep Neural Network (DNN) to be linear in transformed input and feature space improves robustness significantly. We also demonstrate that by augmenting the objective function with Local Lipschitz regularizer boost robustness of the model further. Our method outperforms most sophisticated adversarial training methods and achieves state of the art adversarial accuracy on MNIST, CIFAR10 and SVHN dataset. In this paper, we also propose a novel adversarial image generation method by leveraging Inverse Representation Learning and Linearity aspect of an adversarially trained deep neural network classifier.

In this work we present a clustering technique called \textit{multi-level conformal clustering (MLCC)}. The technique is hierarchical in nature because it can be performed at multiple significance levels which yields greater insight into the data than performing it at just one level. We describe the theoretical underpinnings of MLCC, compare and contrast it with the hierarchical clustering algorithm, and then apply it to real world datasets to assess its performance. There are several advantages to using MLCC over more classical clustering techniques: Once a significance level has been set, MLCC is able to automatically select the number of clusters. Furthermore, thanks to the conformal prediction framework the resulting clustering model has a clear statistical meaning without any assumptions about the distribution of the data. This statistical robustness also allows us to perform clustering and anomaly detection simultaneously. Moreover, due to the flexibility of the conformal prediction framework, our algorithm can be used on top of many other machine learning algorithms.

--T opological data analysis aims to extract topological quantities from data, which tend to focus on the broader global structure of the data rather than local information. The Mapper method, specifically, generalizes clustering methods to identify significant global mathematical structures, which are out of reach of many other approaches. We propose a classifier based on applying the Mapper algorithm to data projected onto a latent space. We obtain the latent space by using PCA or autoencoders. Notably, a classifier based on the Mapper method is immune to any gradient based attack, and improves robustness over traditional CNNs (convolutional neural networks). We report theoretical justification and some numerical experiments that confirm our claims. I NTRODUCTION Deep neural networks [1], [2] are well known to be not robust with respect to input image perturbations, which are designed by adding to images perturbations that are typically non-perceptible by humans [3]-[5]. In this paper we explore opportunities for combining deep learning techniques with a well known topological data analysis (TDA) algorithm - the Mapper algorithm [6], which we use to create classifiers with improved robustness. First, the training data is projected onto a latent space. The latent space in the simplest variant is constructed using PCA components, and we also use nonlinear projections by utilizing various autoencoders [1], [7]-[9]. Then, a discrete graph representation (Mapper) is assigned to the training data projected onto the latent space .

In this paper, the second order convergence of non-convex optimization in the asynchronous stochastic gradient descent (ASGD) algorithm is studied systematically. We investigate the behavior of ASGD near and away from saddle points. Different from the general stochastic gradient descent(SGD), we show that ASGD might return back even if it has escaped the saddle points, yet after staying near a strict saddle point for a long enough time ($O(T)$), ASGD will finally go away from strict saddle points. An inequality is given to describe the process of ASGD to escape saddle points. Using a novel Razumikhin-Lyapunov method, we show the exponential instability of the perturbed gradient dynamics near the strict saddle points and give a more detailed estimation about how the time delay parameter $T$ influences the speed to escape. In particular, we consider the optimization of smooth nonconvex functions, and propose a perturbed asynchronous stochastic gradient descent algorithm with guarantee of convergence to second order stationary points with high probability in $O(1/\epsilon^4)$ iterations. To the best of our knowledge, this is the first work on the second order convergence of asynchronous algorithm.

A keyword spotting (KWS) system determines the existence of, usually predefined, keyword in a continuous speech stream. This paper presents a query-by-example on-device KWS system which is user-specific. The proposed system consists of two main steps: query enrollment and testing. In query enrollment step, phonetic posteriors are output by a small-footprint automatic speech recognition model based on connectionist temporal classification. Using the phonetic-level posteriorgram, hypothesis graph of finite-state transducer (FST) is built, thus can enroll any keywords thus avoiding an out-of-vocabulary problem. In testing, a log-likelihood is scored for input audio using the FST. We propose a threshold prediction method while using the user-specific keyword hypothesis only. The system generates query-specific negatives by rearranging each query utterance in waveform. The threshold is decided based on the enrollment queries and generated negatives. We tested two keywords in English, and the proposed work shows promising performance while preserving simplicity.

Systems of interacting particles or agents have wide applications in many disciplines such as Physics, Chemistry, Biology and Economics. These systems are governed by interaction laws, which are often unknown: estimating them from observation data is a fundamental task that can provide meaningful insights and accurate predictions of the behaviour of the agents. In this paper, we consider the inverse problem of learning interaction laws given data from multiple trajectories, in a nonparametric fashion, when the interaction kernels depend on pairwise distances. We establish a condition for learnability of interaction kernels, and construct estimators that are guaranteed to converge in a suitable $L^2$ space, at the optimal min-max rate for 1-dimensional nonparametric regression. We propose an efficient learning algorithm based on least squares, which can be implemented in parallel for multiple trajectories and is therefore well-suited for the high dimensional, big data regime. Numerical simulations on a variety examples, including opinion dynamics, predator-swarm dynamics and heterogeneous particle dynamics, suggest that the learnability condition is satisfied in models used in practice, and the rate of convergence of our estimator is consistent with the theory. These simulations also suggest that our estimators are robust to noise in the observations, and produce accurate predictions of dynamics in relative large time intervals, even when they are learned from data collected in short time intervals.

Pulmonary embolisms (PE) are known to be one of the leading causes for cardiac-related mortality. Due to inherent variabilities in how PE manifests and the cumbersome nature of manual diagnosis, there is growing interest in leveraging AI tools for detecting PE. In this paper, we build a two-stage detection pipeline that is accurate, computationally efficient, robust to variations in PE types and kernels used for CT reconstruction, and most importantly, does not require dense annotations. Given the challenges in acquiring expert annotations in large-scale datasets, our approach produces state-of-the-art results with very sparse emboli contours (at 10mm slice spacing), while using models with significantly lower number of parameters. We achieve AUC scores of 0.94 on the validation set and 0.85 on the test set of highly severe PEs. Using a large, real-world dataset characterized by complex PE types and patients from multiple hospitals, we present an elaborate empirical study and provide guidelines for designing highly generalizable pipelines.

The runtime performance of modern SAT solvers is deeply connected to the phase transition behavior of CNF formulas. While CNF solving has witnessed significant runtime improvement over the past two decades, the same does not hold for several other classes such as the conjunction of cardinality and XOR constraints, denoted as CARD-XOR formulas. The problem of determining the satisfiability of CARD-XOR formulas is a fundamental problem with a wide variety of applications ranging from discrete integration in the field of artificial intelligence to maximum likelihood decoding in coding theory. The runtime behavior of random CARD-XOR formulas is unexplored in prior work. In this paper, we present the first rigorous empirical study to characterize the runtime behavior of 1-CARD-XOR formulas. We show empirical evidence of a surprising phase-transition that follows a non-linear tradeoff between CARD and XOR constraints.

Dynamic games are an effective paradigm for dealing with the control of multiple interacting actors. Current algorithms for solving these problems either rely on Hamilton-Jacobi-Isaacs (HJI) methods, dynamic programming (DP), differential dynamic programming (DDP), or an iterative best response approach (IBR). The first two approaches have strong theoretical guarantees; however they becomes intractable in high-dimensional real-world applications. The third approach is grounded in the success of iLQR. It is scalable, but it cannot handle constraints. Finally, the iterative best response algorithm is a heuristic approach with unknown convergence properties, and it can suffer from stability and tractability issues. This paper introduces ALGAMES (Augmented Lagrangian GAME-theoretic Solver), a solver that handles trajectory optimization problems with multiple actors and general nonlinear state and input constraints. We evaluate our solver in the context of autonomous driving on scenarios involving numerous vehicles such as ramp merging, overtaking, and lane changing. We present simulation and timing results demonstrating the speed and the ability of the solver to produce efficient, safe, and natural autonomous behaviors.