Parametrized Deep Q-Networks Learning: Reinforcement Learning with Discrete-Continuous Hybrid Action Space

Most existing deep reinforcement learning (DRL) frameworks consider either discrete action space or continuous action space solely. Motivated by applications in computer games, we consider the scenario with discrete-continuous hybrid action space. To handle hybrid action space, previous works either approximate the hybrid space by discretization, or relax it into a continuous set. In this paper, we propose a parametrized deep Q-network (P- DQN) framework for the hybrid action space without approximation or relaxation. Our algorithm combines the spirits of both DQN (dealing with discrete action space) and DDPG (dealing with continuous action space) by seamlessly integrating them. Empirical results on a simulation example, scoring a goal in simulated RoboCup soccer and the solo mode in game King of Glory (KOG) validate the efficiency and effectiveness of our method.

Communication Compression for Decentralized Training

Optimizing distributed learning systems is an art of balancing between computation and communication. There have been two lines of research that try to deal with slower networks: {\em communication compression} for low bandwidth networks, and {\em decentralization} for high latency networks. In this paper, We explore a natural question: {\em can the combination of both techniques lead to a system that is robust to both bandwidth and latency?} Although the system implication of such combination is trivial, the underlying theoretical principle and algorithm design is challenging: unlike centralized algorithms, simply compressing exchanged information, even in an unbiased stochastic way, within the decentralized network would accumulate the error and fail to converge. In this paper, we develop a framework of compressed, decentralized training and propose two different strategies, which we call {\em extrapolation compression} and {\em difference compression}. We analyze both algorithms and prove both converge at the rate of $O(1/\sqrt{nT})$ where $n$ is the number of workers and $T$ is the number of iterations, matching the convergence rate for full precision, centralized training. We validate our algorithms and find that our proposed algorithm outperforms the best of merely decentralized and merely quantized algorithm significantly for networks with {\em both} high latency and low bandwidth.

Marginal Policy Gradients: A Unified Family of Estimators for Bounded Action Spaces with Applications

Many complex domains, such as robotics control and real-time strategy (RTS) games, require an agent to learn a continuous control. In the former, an agent learns a policy over $\mathbb{R}^d$ and in the latter, over a discrete set of actions each of which is parametrized by a continuous parameter. Such problems are naturally solved using policy based reinforcement learning (RL) methods, but unfortunately these often suffer from high variance leading to instability and slow convergence. Unnecessary variance is introduced whenever policies over bounded action spaces are modeled using distributions with unbounded support by applying a transformation $T$ to the sampled action before execution in the environment. Recently, the variance reduced clipped action policy gradient (CAPG) was introduced for actions in bounded intervals, but to date no variance reduced methods exist when the action is a direction, something often seen in RTS games. To this end we introduce the angular policy gradient (APG), a stochastic policy gradient method for directional control. With the marginal policy gradients family of estimators we present a unified analysis of the variance reduction properties of APG and CAPG; our results provide a stronger guarantee than existing analyses for CAPG. Experimental results on a popular RTS game and a navigation task show that the APG estimator offers a substantial improvement over the standard policy gradient.

An Interactive Greedy Approach to Group Sparsity in High Dimensions

Sparsity learning with known grouping structure has received considerable attention due to wide modern applications in high-dimensional data analysis. Although advantages of using group information have been well-studied by shrinkage-based approaches, benefits of group sparsity have not been well-documented for greedy-type methods, which much limits our understanding and use of this important class of methods. In this paper, generalizing from a popular forward-backward greedy approach, we propose a new interactive greedy algorithm for group sparsity learning and prove that the proposed greedy-type algorithm attains the desired benefits of group sparsity under high dimensional settings. An estimation error bound refining other existing methods and a guarantee for group support recovery are also established simultaneously. In addition, we incorporate a general M-estimation framework and introduce an interactive feature to allow extra algorithm flexibility without compromise in theoretical properties. The promising use of our proposal is demonstrated through numerical evaluations including a real industrial application in human activity recognition at home. Supplementary materials for this article are available online.

Asynchronous Decentralized Parallel Stochastic Gradient Descent

Most commonly used distributed machine learning systems are either synchronous or centralized asynchronous. Synchronous algorithms like AllReduce-SGD perform poorly in a heterogeneous environment, while asynchronous algorithms using a parameter server suffer from 1) communication bottleneck at parameter servers when workers are many, and 2) significantly worse convergence when the traffic to parameter server is congested. Can we design an algorithm that is robust in a heterogeneous environment, while being communication efficient and maintaining the best-possible convergence rate? In this paper, we propose an asynchronous decentralized stochastic gradient decent algorithm (AD-PSGD) satisfying all above expectations. Our theoretical analysis shows AD-PSGD converges at the optimal $O(1/\sqrt{K})$ rate as SGD and has linear speedup w.r.t. number of workers. Empirically, AD-PSGD outperforms the best of decentralized parallel SGD (D-PSGD), asynchronous parallel SGD (A-PSGD), and standard data parallel SGD (AllReduce-SGD), often by orders of magnitude in a heterogeneous environment. When training ResNet-50 on ImageNet with up to 128 GPUs, AD-PSGD converges (w.r.t epochs) similarly to the AllReduce-SGD, but each epoch can be up to 4-8X faster than its synchronous counterparts in a network-sharing HPC environment. To the best of our knowledge, AD-PSGD is the first asynchronous algorithm that achieves a similar epoch-wise convergence rate as AllReduce-SGD, at an over 100-GPU scale.

Stochastically Controlled Stochastic Gradient for the Convex and Non-convex Composition problem

In this paper, we consider the convex and non-convex composition problem with the structure $\frac{1}{n}\sum\nolimits_{i = 1}^n {{F_i}( {G( x )} )}$, where $G( x )=\frac{1}{n}\sum\nolimits_{j = 1}^n {{G_j}( x )} $ is the inner function, and $F_i(\cdot)$ is the outer function. We explore the variance reduction based method to solve the composition optimization. Due to the fact that when the number of inner function and outer function are large, it is not reasonable to estimate them directly, thus we apply the stochastically controlled stochastic gradient (SCSG) method to estimate the gradient of the composition function and the value of the inner function. The query complexity of our proposed method for the convex and non-convex problem is equal to or better than the current method for the composition problem. Furthermore, we also present the mini-batch version of the proposed method, which has the improved the query complexity with related to the size of the mini-batch.

End-to-End Learning of Energy-Constrained Deep Neural Networks

Deep Neural Networks (DNN) are increasingly deployed in highly energy-constrained environments such as autonomous drones and wearable devices while at the same time must operate in real-time. Therefore, reducing the energy consumption has become a major design consideration in DNN training. This paper proposes the first end-to-end DNN training framework that provides quantitative energy guarantees. The key idea is to formulate the DNN training as an optimization problem in which the energy budget imposes a previously unconsidered optimization constraint. We integrate the quantitative DNN energy estimation into the DNN training process to assist the constraint optimization. We prove that an approximate algorithm can be used to efficiently solve the optimization problem. Compared to the best prior energy-saving techniques, our framework trains DNNs that provide higher accuracies under same or lower energy budgets.

GESF: A Universal Discriminative Mapping Mechanism for Graph Representation Learning

Graph embedding is a central problem in social network analysis and many other applications, aiming to learn the vector representation for each node. While most existing approaches need to specify the neighborhood and the dependence form to the neighborhood, which may significantly degrades the flexibility of representation, we propose a novel graph node embedding method (namely GESF) via the set function technique. Our method can 1) learn an arbitrary form of representation function from neighborhood, 2) automatically decide the significance of neighbors at different distances, and 3) be applied to heterogeneous graph embedding, which may contain multiple types of nodes. Theoretical guarantee for the representation capability of our method has been proved for general homogeneous and heterogeneous graphs and evaluation results on benchmark data sets show that the proposed GESF outperforms the state-of-the-art approaches on producing node vectors for classification tasks.

Parallel Computation of PDFs on Big Spatial Data Using Spark

We consider big spatial data, which is typically produced in scientific areas such as geological or seismic interpretation. The spatial data can be produced by observation (e.g. using sensors or soil instrument) or numerical simulation programs and correspond to points that represent a 3D soil cube area. However, errors in signal processing and modeling create some uncertainty, and thus a lack of accuracy in identifying geological or seismic phenomenons. Such uncertainty must be carefully analyzed. To analyze uncertainty, the main solution is to compute a Probability Density Function (PDF) of each point in the spatial cube area. However, computing PDFs on big spatial data can be very time consuming (from several hours to even months on a parallel computer). In this paper, we propose a new solution to efficiently compute such PDFs in parallel using Spark, with three methods: data grouping, machine learning prediction and sampling. We evaluate our solution by extensive experiments on different computer clusters using big data ranging from hundreds of GB to several TB. The experimental results show that our solution scales up very well and can reduce the execution time by a factor of 33 (in the order of seconds or minutes) compared with a baseline method.

D$^2$: Decentralized Training over Decentralized Data

While training a machine learning model using multiple workers, each of which collects data from their own data sources, it would be most useful when the data collected from different workers can be {\em unique} and {\em different}. Ironically, recent analysis of decentralized parallel stochastic gradient descent (D-PSGD) relies on the assumption that the data hosted on different workers are {\em not too different}. In this paper, we ask the question: {\em Can we design a decentralized parallel stochastic gradient descent algorithm that is less sensitive to the data variance across workers?} In this paper, we present D$^2$, a novel decentralized parallel stochastic gradient descent algorithm designed for large data variance \xr{among workers} (imprecisely, "decentralized" data). The core of D$^2$ is a variance blackuction extension of the standard D-PSGD algorithm, which improves the convergence rate from $O\left({\sigma \over \sqrt{nT}} + {(n\zeta^2)^{\frac{1}{3}} \over T^{2/3}}\right)$ to $O\left({\sigma \over \sqrt{nT}}\right)$ where $\zeta^{2}$ denotes the variance among data on different workers. As a result, D$^2$ is robust to data variance among workers. We empirically evaluated D$^2$ on image classification tasks where each worker has access to only the data of a limited set of labels, and find that D$^2$ significantly outperforms D-PSGD.