Exploration in environments with continuous control and sparse rewards remains a key challenge in reinforcement learning (RL). Recently, surprise has been used as an intrinsic reward that encourages systematic and efficient exploration. We introduce a new definition of surprise and its RL implementation named Variational Assorted Surprise Exploration (VASE). VASE uses a Bayesian neural network as a model of the environment dynamics and is trained using variational inference, alternately updating the accuracy of the agent's model and policy. Our experiments show that in continuous control sparse reward environments VASE outperforms other surprise-based exploration techniques.

In this paper, we propose and analyze SPARQ-SGD, which is an event-triggered and compressed algorithm for decentralized training of large-scale machine learning models. Each node can locally compute a condition (event) which triggers a communication where quantized and sparsified local model parameters are sent. In SPARQ-SGD each node takes at least a fixed number ($H$) of local gradient steps and then checks if the model parameters have significantly changed compared to its last update; it communicates further compressed model parameters only when there is a significant change, as specified by a (design) criterion. We prove that the SPARQ-SGD converges as $O(\frac{1}{nT})$ and $O(\frac{1}{\sqrt{nT}})$ in the strongly-convex and non-convex settings, respectively, demonstrating that such aggressive compression, including event-triggered communication, model sparsification and quantization does not affect the overall convergence rate as compared to uncompressed decentralized training; thereby theoretically yielding communication efficiency for "free". We evaluate SPARQ-SGD over real datasets to demonstrate significant amount of savings in communication over the state-of-the-art.

User identity linkage is a task of recognizing the identities of the same user across different social networks (SN). Previous works tackle this problem via estimating the pairwise similarity between identities from different SN, predicting the label of identity pairs or selecting the most relevant identity pair based on the similarity scores. However, most of these methods ignore the results of previously matched identities, which could contribute to the linkage in following matching steps. To address this problem, we convert user identity linkage into a sequence decision problem and propose a reinforcement learning model to optimize the linkage strategy from the global perspective. Our method makes full use of both the social network structure and the history matched identities, and explores the long-term influence of current matching on subsequent decisions. We conduct experiments on different types of datasets, the results show that our method achieves better performance than other state-of-the-art methods.

Training deep neural networks (DNN) is expensive in terms of computational power and the amount of necessary labeled training data. Thus, deep learning models constitute business value for data owners. Watermarking of deep neural networks can enable their tracing once released by a data owner. In this paper we define and formalize white-box watermarking algorithms for DNNs, where the data owner needs white-box access to the model to extract the watermark. White-box watermarking algorithms have the advantage that they do not impact the accuracy of the watermarked model. We demonstrate a new property inference attack using a DNN that can detect watermarking by any existing, manually designed algorithms regardless of training dataset and model architecture. We then propose the first white-box DNN watermarking algorithm that is undetectable by the property inference attack. We further extend the capacity and robustness of the watermark. Unlike prior watermarking schemes which restrict the content of watermark message to short binary strings, our new scheme largely increase the capacity and flexibility of the embedded watermark message. Experiments show that our new white-box watermarking algorithm does not impact accuracy, is undetectable and robust against moderate model transformation attacks.

Energy-based models (EBMs) are powerful probabilistic models, but suffer from intractable sampling and density evaluation due to the partition function. As a result, inference in EBMs relies on approximate sampling algorithms, leading to a mismatch between the model and inference. Motivated by this, we consider the sampler-induced distribution as the model of interest and maximize the likelihood of this model. This yields a class of energy-inspired models (EIMs) that incorporate learned energy functions while still providing exact samples and tractable log-likelihood lower bounds. We describe and evaluate three instantiations of such models based on truncated rejection sampling, self-normalized importance sampling, and Hamiltonian importance sampling. These models outperform or perform comparably to the recently proposed Learned Accept/Reject Sampling algorithm and provide new insights on ranking Noise Contrastive Estimation and Contrastive Predictive Coding. Moreover, EIMs allow us to generalize a recent connection between multi-sample variational lower bounds and auxiliary variable variational inference. We show how recent variational bounds can be unified with EIMs as the variational family.

Automatic machine learning (AutoML) is an area of research aimed at automating machine learning (ML) activities that currently require human experts. One of the most challenging tasks in this field is the automatic generation of end-to- end ML pipelines: combining multiple types of ML algorithms into a single architecture used for end-to-end analysis of previously-unseen data. This task has two challenging aspects: the first is the need to explore a large search space of algorithms and pipeline architectures. The second challenge is the computational cost of training and evaluating multiple pipelines. In this study we present DeepLine, a reinforcement learning based approach for automatic pipeline generation. Our proposed approach utilizes an efficient representation of the search space and leverages past knowledge gained from previously-analyzed datasets to make the problem more tractable. Additionally, we propose a novel hierarchical-actions algorithm that serves as a plugin, mediating the environment-agent interaction in deep reinforcement learning problems. The plugin significantly speeds up the training process of our model. Evaluation on 56 datasets shows that DeepLine outperforms state-of-the-art approaches both in accuracy and in computational cost.

Continual learning aims to improve the ability of modern learning systems to deal with non-stationary distributions, typically by attempting to learn a series of tasks sequentially. Prior art in the field has largely considered supervised or reinforcement learning tasks, and often assumes full knowledge of task labels and boundaries. In this work, we propose an approach (CURL) to tackle a more general problem that we will refer to as unsupervised continual learning. The focus is on learning representations without any knowledge about task identity, and we explore scenarios when there are abrupt changes between tasks, smooth transitions from one task to another, or even when the data is shuffled. The proposed approach performs task inference directly within the model, is able to dynamically expand to capture new concepts over its lifetime, and incorporates additional rehearsal-based techniques to deal with catastrophic forgetting. We demonstrate the efficacy of CURL in an unsupervised learning setting with MNIST and Omniglot, where the lack of labels ensures no information is leaked about the task. Further, we demonstrate strong performance compared to prior art in an i.i.d setting, or when adapting the technique to supervised tasks such as incremental class learning.

Fairness is essential for human society, contributing to stability and productivity. Similarly, fairness is also the key for many multi-agent systems. Taking fairness into multi-agent learning could help multi-agent systems become both efficient and stable. However, learning efficiency and fairness simultaneously is a complex, multi-objective, joint-policy optimization. To tackle these difficulties, we propose FEN, a novel hierarchical reinforcement learning model. We first decompose fairness for each agent and propose fair-efficient reward that each agent learns its own policy to optimize. To avoid multi-objective conflict, we design a hierarchy consisting of a controller and several sub-policies, where the controller maximizes the fair-efficient reward by switching among the sub-policies that provides diverse behaviors to interact with the environment. FEN can be trained in a fully decentralized way, making it easy to be deployed in real-world applications. Empirically, we show that FEN easily learns both fairness and efficiency and significantly outperforms baselines in a variety of multi-agent scenarios.

Black Box Machine Learning models leak information about the proprietary model parameters and architecture, both through side channels and output predictions. An adversary can thus, exploit this leakage to reconstruct a substitute architecture similar to the target model, violating the model privacy and Intellectual Property. However, all such attacks, infer a subset of the target model attributes and identifying the rest of the architecture and parameters (optimally) is a search problem. Extracting the exact target model is not possible owing to the uncertainty in the inference attack outputs and stochastic nature of the training process. In this work, we propose a probabilistic framework, Airavata, to estimate the leakage in such model extraction attacks. Specifically, we use Bayesian Networks to capture the uncertainty, under the subjective notion of probability, in estimating the target model attributes using various model extraction attacks. We experimentally validate the model under different adversary assumptions commonly adopted by various model extraction attacks to reason about the attack efficacy. Further, this provides a practical approach of inferring actionable knowledge about extracting black box models and identify the best combination of attacks which maximise the knowledge extracted (information leaked) from the target model.

We introduce agents that use object-oriented reasoning to consider alternate states of the world in order to more quickly find solutions to problems. Specifically, a hierarchical controller directs a low-level agent to behave as if objects in the scene were added, deleted, or modified. The actions taken by the controller are defined over a graph-based representation of the scene, with actions corresponding to adding, deleting, or editing the nodes of a graph. We present preliminary results on three environments, demonstrating that our approach can achieve similar levels of reward as non-hierarchical agents, but with better data efficiency.