Modern recommender systems (RS) work by processing a number of signals that can be inferred from large sets of user-item interaction data. The main signal to analyze stems from the raw matrix that represents interactions. However, we can increase the performance of RS by considering other kinds of signals like the context of interactions, which could be, for example, the time or date of the interaction, the user location, or sequential data corresponding to the historical interactions of the user with the system. These complex, context-based interaction signals are characterized by a rich relational structure that can be represented by a multi-partite graph. Graph Convolutional Networks (GCNs) have been used successfully in collaborative filtering with simple user-item interaction data. In this work, we generalize the use of GCNs for N-partite graphs by considering N multiple context dimensions and propose a simple way for their seamless integration in modern deep learning RS architectures. More specifically, we define a graph convolutional embedding layer for N-partite graphs that processes user-item-context interactions, and constructs node embeddings by leveraging their relational structure. Experiments on several datasets from recommender systems to drug re-purposing show the benefits of the introduced GCN embedding layer by measuring the performance of different context-enriched tasks.
Seeking to improve model generalization, we consider a new approach based on distributionally robust learning (DRL) that applies stochastic gradient descent to the outer minimization problem. Our algorithm efficiently estimates the gradient of the inner maximization problem through multi-level Monte Carlo randomization. Leveraging theoretical results that shed light on why standard gradient estimators fail, we establish the optimal parameterization of the gradient estimators of our approach that balances a fundamental tradeoff between computation time and statistical variance. Numerical experiments demonstrate that our DRL approach yields significant benefits over previous work.
Detection of nocturnal seizures in epilepsy patients is essential, both for the quick management of the seizure complications, and for the assessment of the ongoing seizure treatment. Traditional seizure detection products (e.g., wearables), however, are either very costly, uncomfortable, or unreliable. In this paper, we then propose to utilize everyday WiFi signals for robust, fast, and non-invasive detection of nocturnal seizures. We first present a new and rigorous mathematical characterization for the spectral content/bandwidth of the WiFi signal, measured on a WiFi device placed near a sleeping patient, during different kinds of sleep motions: seizures, normal movements (e.g. posture adjustments), and breathing. Based on this mathematical modeling, we propose a novel pipeline for processing the received WiFi signals to robustly detect all nocturnal non-breathing movements, and then classify them into normal body movements or seizures. In order to validate this, we carry out extensive experiments in 7 different typical bedroom locations, where a set of 20 actors simulate the state of having seizures (total of 260 instances), as well as normal sleep movements (total of 410 instances). Our proposed system detects 93.85% of the seizures with a mean response time of only 5.69 seconds since the onset of the seizure. Moreover, our proposed system achieves a probability of false alarm of only 0.0097, when classifying normal sleep movements. Overall, our new mathematical modeling and experimental results show the great potential the ubiquitous WiFi signals have for detecting nocturnal seizures, which can provide better support for epilepsy patients and their caregivers.
We study the emergence of chaotic behavior of Follow-the-Regularized Leader (FoReL) dynamics in games. We focus on the effects of increasing the population size or the scale of costs in congestion games, and generalize recent results on unstable, chaotic behaviors in the Multiplicative Weights Update dynamics to a much larger class of FoReL dynamics. We establish that, even in simple linear non-atomic congestion games with two parallel links and any fixed learning rate, unless the game is fully symmetric, increasing the population size or the scale of costs causes learning dynamics to become unstable and eventually chaotic, in the sense of Li-Yorke and positive topological entropy. Furthermore, we show the existence of novel non-standard phenomena such as the coexistence of stable Nash equilibria and chaos in the same game. We also observe the simultaneous creation of a chaotic attractor as another chaotic attractor gets destroyed. Lastly, although FoReL dynamics can be strange and non-equilibrating, we prove that the time average still converges to an exact equilibrium for any choice of learning rate and any scale of costs.
We introduce a framework for Compositional Imitation Learning and Execution (CompILE) of hierarchically-structured behavior. CompILE learns reusable, variable-length segments of behavior from demonstration data using a novel unsupervised, fully-differentiable sequence segmentation module. These learned behaviors can then be re-composed and executed to perform new tasks. At training time, CompILE auto-encodes observed behavior into a sequence of latent codes, each corresponding to a variable-length segment in the input sequence. Once trained, our model generalizes to sequences of longer length and from environment instances not seen during training. We evaluate our model in a challenging 2D multi-task environment and show that CompILE can find correct task boundaries and event encodings in an unsupervised manner without requiring annotated demonstration data. Latent codes and associated behavior policies discovered by CompILE can be used by a hierarchical agent, where the high-level policy selects actions in the latent code space, and the low-level, task-specific policies are simply the learned decoders. We found that our agent could learn given only sparse rewards, where agents without task-specific policies struggle.
In this paper, we study Markov Decision Processes (MDPs) with self-triggered strategies, where the idea of self-triggered control is extended to more generic MDP models. This extension broadens the application of self-triggering policies to a broader range of systems. We study the co-design problems of the control policy and the triggering policy to optimize two pre-specified cost criteria. The first cost criterion is introduced by incorporating a pre-specified update penalty into the traditional MDP cost criteria to reduce the use of communication resources. Under this criteria, a novel dynamic programming (DP) equation called DP equation with optimized lookahead to proposed to solve for the self-triggering policy under this criteria. The second self-triggering policy is to maximize the triggering time while still guaranteeing a pre-specified level of sub-optimality. Theoretical underpinnings are established for the computation and implementation of both policies. Through a gridworld numerical example, we illustrate the two policies' effectiveness in reducing sources consumption and demonstrate the trade-offs between resource consumption and system performance.
3D scene representation for robot manipulation should capture three key object properties: permanency -- objects that become occluded over time continue to exist; amodal completeness -- objects have 3D occupancy, even if only partial observations are available; spatiotemporal continuity -- the movement of each object is continuous over space and time. In this paper, we introduce 3D Dynamic Scene Representation (DSR), a 3D volumetric scene representation that simultaneously discovers, tracks, reconstructs objects, and predicts their dynamics while capturing all three properties. We further propose DSR-Net, which learns to aggregate visual observations over multiple interactions to gradually build and refine DSR. Our model achieves state-of-the-art performance in modeling 3D scene dynamics with DSR on both simulated and real data. Combined with model predictive control, DSR-Net enables accurate planning in downstream robotic manipulation tasks such as planar pushing. Video is available at https://youtu.be/GQjYG3nQJ80.
Some deep convolutional neural networks were proposed for time-series classification and class imbalanced problems. However, those models performed degraded and even failed to recognize the minority class of an imbalanced temporal sequences dataset. Minority samples would bring troubles for temporal deep learning classifiers due to the equal treatments of majority and minority class. Until recently, there were few works applying deep learning on imbalanced time-series classification (ITSC) tasks. Here, this paper aimed at tackling ITSC problems with deep learning. An adaptive cost-sensitive learning strategy was proposed to modify temporal deep learning models. Through the proposed strategy, classifiers could automatically assign misclassification penalties to each class. In the experimental section, the proposed method was utilized to modify five neural networks. They were evaluated on a large volume, real-life and imbalanced time-series dataset with six metrics. Each single network was also tested alone and combined with several mainstream data samplers. Experimental results illustrated that the proposed cost-sensitive modified networks worked well on ITSC tasks. Compared to other methods, the cost-sensitive convolution neural network and residual network won out in the terms of all metrics. Consequently, the proposed cost-sensitive learning strategy can be used to modify deep learning classifiers from cost-insensitive to cost-sensitive. Those cost-sensitive convolutional networks can be effectively applied to address ITSC issues.
Federated learning (FL) utilizes edge computing devices to collaboratively train a shared model while each device can fully control its local data access. Generally, FL techniques focus on learning model on independent and identically distributed (iid) dataset and cannot achieve satisfiable performance on non-iid datasets (e.g. learning a multi-class classifier but each client only has a single class dataset). Some personalized approaches have been proposed to mitigate non-iid issues. However, such approaches cannot handle underlying data distribution shift, namely data distribution skew, which is quite common in real scenarios (e.g. recommendation systems learn user behaviors which change over time). In this work, we provide a solution to the challenge by leveraging smart-contract with federated learning to build optimized, personalized deep learning models. Specifically, our approach utilizes smart contract to reach consensus among distributed trainers on the optimal weights of personalized models. We conduct experiments across multiple models (CNN and MLP) and multiple datasets (MNIST and CIFAR-10). The experimental results demonstrate that our personalized learning models can achieve better accuracy and faster convergence compared to classic federated and personalized learning. Compared with the model given by baseline FedAvg algorithm, the average accuracy of our personalized learning models is improved by 2% to 20%, and the convergence rate is about 2$\times$ faster. Moreover, we also illustrate that our approach is secure against recent attack on distributed learning.
A growing number of machine learning architectures, such as Generative Adversarial Networks, rely on the design of games which implement a desired functionality via a Nash equilibrium. In practice these games have an implicit complexity (e.g. from underlying datasets and the deep networks used) that makes directly computing a Nash equilibrium impractical or impossible. For this reason, numerous learning algorithms have been developed with the goal of iteratively converging to a Nash equilibrium. Unfortunately, the dynamics generated by the learning process can be very intricate and instances of training failure hard to interpret. In this paper we show that, in a strong sense, this dynamic complexity is inherent to games. Specifically, we prove that replicator dynamics, the continuous-time analogue of Multiplicative Weights Update, even when applied in a very restricted class of games -- known as finite matrix games -- is rich enough to be able to approximate arbitrary dynamical systems. Our results are positive in the sense that they show the nearly boundless dynamic modelling capabilities of current machine learning practices, but also negative in implying that these capabilities may come at the cost of interpretability. As a concrete example, we show how replicator dynamics can effectively reproduce the well-known strange attractor of Lonrenz dynamics (the "butterfly effect") while achieving no regret.