Adversarial training is the most successful empirical method, to increase the robustness of neural networks against adversarial attacks yet. Unfortunately, this higher robustness is accompanied by considerably higher computational complexity. To date, only adversarial training with expensive multi-step adversarial attacks like Projected Gradient Descent (PGD) proved effective against equally strong attacks. In this paper, we present two ideas that combined enable adversarial training with the computationally less expensive Fast Gradient Sign Method (FGSM). First, we add uniform noise to the initial data point of the FGSM attack, which creates a wider variety of stronger adversaries. Further, we add a learnable regularization step prior to the neural network called Stochastic Augmentation Layer (SAL). Inputs propagated trough the SAL are resampled from a Gaussian distribution. The randomness of the resampling at inference time makes it more complicated for the attacker to construct an adversarial example since the outcome of the model is not known in advance. We show that noise injection in conjunction with FGSM adversarial training achieves comparable results to adversarial training with PGD while being orders of magnitude faster. Moreover, we show superior results in comparison to PGD-based training when combining noise injection and SAL.

Human affect recognition is an essential part of natural human-computer interaction. However, current methods are still in their infancy, especially for in-the-wild data. In this work, we introduce our submission to the Affective Behavior Analysis in-the-wild (ABAW) 2020 competition. We propose a two-stream aural-visual analysis model to recognize affective behavior from videos. Audio and image streams are first processed separately and fed into a convolutional neural network. Instead of applying recurrent architectures for temporal analysis we only use temporal convolutions. Furthermore, the model is given access to additional features extracted during face-alignment. At training time, we exploit correlations between different emotion representations to improve performance. Our model achieves promising results on the challenging Aff-Wild2 database.

Quality-Diversity (QD) algorithms, and MAP-Elites (ME) in particular, have proven very useful for a broad range of applications including enabling real robots to recover quickly from joint damage, solving strongly deceptive maze tasks or evolving robot morphologies to discover new gaits. However, present implementations of MAP-Elites and other QD algorithms seem to be limited to low-dimensional controllers with far fewer parameters than modern deep neural network models. In this paper, we propose to leverage the efficiency of Evolution Strategies (ES) to scale MAP-Elites to high-dimensional controllers parameterized by large neural networks. We design and evaluate a new hybrid algorithm called MAP-Elites with Evolution Strategies (ME-ES) for post-damage recovery in a difficult high-dimensional control task where traditional ME fails. Additionally,we show that ME-ES performs efficient exploration, on par with state-of-the-art exploration algorithms in high-dimensional control tasks with strongly deceptive rewards.

Learning long-range behaviors on complex high-dimensional agents is a fundamental problem in robot learning. For such tasks, we argue that transferring learned information from a morphologically simpler agent can massively improve the sample efficiency of a more complex one. To this end, we propose a hierarchical decoupling of policies into two parts: an independently learned low-level policy and a transferable high-level policy. To remedy poor transfer performance due to mismatch in morphologies, we contribute two key ideas. First, we show that incentivizing a complex agent's low-level to imitate a simpler agent's low-level significantly improves zero-shot high-level transfer. Second, we show that KL-regularized training of the high level stabilizes learning and prevents mode-collapse. Finally, on a suite of publicly released navigation and manipulation environments, we demonstrate the applicability of hierarchical transfer on long-range tasks across morphologies. Our code and videos can be found at https://sites.google.com/berkeley.edu/morphology-transfer.

The endeavor to understand the brain involves multiple collaborating research fields. Classically, synaptic plasticity rules derived by theoretical neuroscientists are evaluated in isolation on pattern classification tasks. This contrasts with the biological brain which purpose is to control a body in closed-loop. This paper contributes to bringing the fields of computational neuroscience and robotics closer together by integrating open-source software components from these two fields. The resulting framework allows to evaluate the validity of biologically-plausibe plasticity models in closed-loop robotics environments. We demonstrate this framework to evaluate Synaptic Plasticity with Online REinforcement learning (SPORE), a reward-learning rule based on synaptic sampling, on two visuomotor tasks: reaching and lane following. We show that SPORE is capable of learning to perform policies within the course of simulated hours for both tasks. Provisional parameter explorations indicate that the learning rate and the temperature driving the stochastic processes that govern synaptic learning dynamics need to be regulated for performance improvements to be retained. We conclude by discussing the recent deep reinforcement learning techniques which would be beneficial to increase the functionality of SPORE on visuomotor tasks.

Current deep reinforcement learning approaches incorporate minimal prior knowledge about the environment, limiting computational and sample efficiency. We incorporate a few object-based priors that humans are known to use: "Infants divide perceptual arrays into units that move as connected wholes, that move separately from one another, that tend to maintain their size and shape over motion, and that tend to act upon each other only on contact" [Spelke]. We propose a probabilistic object-based model of environments and use human object priors to develop an efficient self-supervised algorithm for maximum likelihood estimation of the model parameters from observations and for inferring objects directly from the perceptual stream. We then use object features and incorporate object-contact priors to improve the sample efficiency our object-based RL agent.We evaluate our approach on a subset of the Atari benchmarks, and learn up to four orders of magnitude faster than the standard deep Q-learning network, rendering rapid desktop experiments in this domain feasible. To our knowledge, our system is the first to learn any Atari task in fewer environment interactions than humans.

Despite recent success of deep network-based Reinforcement Learning (RL), it remains elusive to achieve human-level efficiency in learning novel tasks. While previous efforts attempt to address this challenge using meta-learning strategies, they typically suffer from sampling inefficiency with on-policy RL algorithms or meta-overfitting with off-policy learning. In this work, we propose a novel meta-RL strategy to address those limitations. In particular, we decompose the meta-RL problem into three sub-tasks, task-exploration, task-inference and task-fulfillment, instantiated with two deep network agents and a task encoder. During meta-training, our method learns a task-conditioned actor network for task-fulfillment, an explorer network with a self-supervised reward shaping that encourages task-informative experiences in task-exploration, and a context-aware graph-based task encoder for task inference. We validate our approach with extensive experiments on several public benchmarks and the results show that our algorithm effectively performs exploration for task inference, improves sample efficiency during both training and testing, and mitigates the meta-overfitting problem.

Task 1 of the DSTC8-track1 challenge aims to develop an end-to-end multi-domain dialogue system to accomplish complex users' goals under tourist information desk settings. This paper describes our submitted solution, Hierarchical Context Enhanced Dialogue System (HCEDS), for this task. The main motivation of our system is to comprehensively explore the potential of hierarchical context for sufficiently understanding complex dialogues. More specifically, we apply BERT to capture token-level information and employ the attention mechanism to capture sentence-level information. The results listed in the leaderboard show that our system achieves first place in automatic evaluation and the second place in human evaluation.

Modeling how human moves on the space is useful for policy-making in transportation, public safety, and public health. The human movements can be viewed as a dynamic process that human transits between states (e.g., locations) over time. In the human world where both intelligent agents like humans or vehicles with human drivers play an important role, the states of agents mostly describe human activities, and the state transition is influenced by both the human decisions and physical constraints from the real-world system (e.g., agents need to spend time to move over a certain distance). Therefore, the modeling of state transition should include the modeling of the agent's decision process and the physical system dynamics. In this paper, we propose to model state transition in human movement through learning decision model and integrating system dynamics. In experiments on real-world datasets, we demonstrate that the proposed method can achieve superior performance against the state-of-the-art methods in predicting the next state and generating long-term future states.

On behalf of the entire community, we are proud to announce Kubeflow 1.0, our first major release. Kubeflow was open sourced at Kubecon USA in December 2017, and during the last two years the Kubeflow Project has grown beyond our wildest expectations. There are now hundreds of contributors from over 30 participating organizations. Kubeflow's goal is to make it easy for machine learning (ML) engineers and data scientists to leverage cloud assets (public or on-premise) for ML workloads. You can use Kubeflow on any Kubernetes-conformant cluster.