Jiechao Xiong

Lei Han, Jiechao Xiong, Peng Sun et al.

StarCraft, one of the most difficult esport games with long-standing history of professional tournaments, has attracted generations of players and fans, and also, intense attentions in artificial intelligence research. Recently, Google's DeepMind announced AlphaStar, a grandmaster level AI in StarCraft II that can play with humans using comparable action space and operations. In this paper, we introduce a new AI agent, named TStarBot-X, that is trained under orders of less computations and can play competitively with expert human players. TStarBot-X takes advantage of important techniques introduced in AlphaStar, and also benefits from substantial innovations including new league training methods, novel multi-agent roles, rule-guided policy search, stabilized policy improvement, lightweight neural network architecture, and importance sampling in imitation learning, etc. We show that with orders of less computation scale, a faithful reimplementation of AlphaStar's methods can not succeed and the proposed techniques are necessary to ensure TStarBot-X's competitive performance. We reveal all technical details that are complementary to those mentioned in AlphaStar, showing the most sensitive parts in league training, reinforcement learning and imitation learning that affect the performance of the agents. Most importantly, this is an open-sourced study that all codes and resources (including the trained model parameters) are publicly accessible via \url{https://github.com/tencent-ailab/tleague_projpage}. We expect this study could be beneficial for both academic and industrial future research in solving complex problems like StarCraft, and also, might provide a sparring partner for all StarCraft II players and other AI agents.

Peng Sun, Jiechao Xiong, Lei Han et al.

Competitive Self-Play (CSP) based Multi-Agent Reinforcement Learning (MARL) has shown phenomenal breakthroughs recently. Strong AIs are achieved for several benchmarks, including Dota 2, Glory of Kings, Quake III, StarCraft II, to name a few. Despite the success, the MARL training is extremely data thirsty, requiring typically billions of (if not trillions of) frames be seen from the environment during training in order for learning a high performance agent. This poses non-trivial difficulties for researchers or engineers and prevents the application of MARL to a broader range of real-world problems. To address this issue, in this manuscript we describe a framework, referred to as TLeague, that aims at large-scale training and implements several main-stream CSP-MARL algorithms. The training can be deployed in either a single machine or a cluster of hybrid machines (CPUs and GPUs), where the standard Kubernetes is supported in a cloud native manner. TLeague achieves a high throughput and a reasonable scale-up when performing distributed training. Thanks to the modular design, it is also easy to extend for solving other multi-agent problems or implementing and verifying MARL algorithms. We present experiments over StarCraft II, ViZDoom and Pommerman to show the efficiency and effectiveness of TLeague. The code is open-sourced and available at https://github.com/tencent-ailab/tleague_projpage

Qing Wang, Yingru Li, Jiechao Xiong et al.

In deep reinforcement learning, policy optimization methods need to deal with issues such as function approximation and the reuse of off-policy data. Standard policy gradient methods do not handle off-policy data well, leading to premature convergence and instability. This paper introduces a method to stabilize policy optimization when off-policy data are reused. The idea is to include a Bregman divergence between the behavior policy that generates the data and the current policy to ensure small and safe policy updates with off-policy data. The Bregman divergence is calculated between the state distributions of two policies, instead of only on the action probabilities, leading to a divergence augmentation formulation. Empirical experiments on Atari games show that in the data-scarce scenario where the reuse of off-policy data becomes necessary, our method can achieve better performance than other state-of-the-art deep reinforcement learning algorithms.

Hanlin Yang, Jian Yao, Weiming Liu et al.

Recovering a spectrum of diverse policies from a set of expert trajectories is an important research topic in imitation learning. After determining a latent style for a trajectory, previous diverse policies recovering methods usually employ a vanilla behavioral cloning learning objective conditioned on the latent style, treating each state-action pair in the trajectory with equal importance. Based on an observation that in many scenarios, behavioral styles are often highly relevant with only a subset of state-action pairs, this paper presents a new principled method in diverse polices recovery. In particular, after inferring or assigning a latent style for a trajectory, we enhance the vanilla behavioral cloning by incorporating a weighting mechanism based on pointwise mutual information. This additional weighting reflects the significance of each state-action pair's contribution to learning the style, thus allowing our method to focus on state-action pairs most representative of that style. We provide theoretical justifications for our new objective, and extensive empirical evaluations confirm the effectiveness of our method in recovering diverse policies from expert data.

Hao Sun, Ziping Xu, Yuhang Song et al.

Policy gradient (PG) algorithms have been widely used in reinforcement learning (RL). However, PG algorithms rely on exploiting the value function being learned with the first-order update locally, which results in limited sample efficiency. In this work, we propose an alternative method called Zeroth-Order Supervised Policy Improvement (ZOSPI). ZOSPI exploits the estimated value function $Q$ globally while preserving the local exploitation of the PG methods based on zeroth-order policy optimization. This learning paradigm follows Q-learning but overcomes the difficulty of efficiently operating argmax in continuous action space. It finds max-valued action within a small number of samples. The policy learning of ZOSPI has two steps: First, it samples actions and evaluates those actions with a learned value estimator, and then it learns to perform the action with the highest value through supervised learning. We further demonstrate such a supervised learning framework can learn multi-modal policies. Experiments show that ZOSPI achieves competitive results on the continuous control benchmarks with a remarkable sample efficiency.

Qing Wang, Jiechao Xiong, Lei Han et al.

We introduce Arena, a toolkit for multi-agent reinforcement learning (MARL) research. In MARL, it usually requires customizing observations, rewards and actions for each agent, changing cooperative-competitive agent-interaction, and playing with/against a third-party agent, etc. We provide a novel modular design, called Interface, for manipulating such routines in essentially two ways: 1) Different interfaces can be concatenated and combined, which extends the OpenAI Gym Wrappers concept to MARL scenarios. 2) During MARL training or testing, interfaces can be embedded in either wrapped OpenAI Gym compatible Environments or raw environment compatible Agents. We offer off-the-shelf interfaces for several popular MARL platforms, including StarCraft II, Pommerman, ViZDoom, Soccer, etc. The interfaces effectively support self-play RL and cooperative-competitive hybrid MARL. Also, Arena can be conveniently extended to your own favorite MARL platform.

Jiechao Xiong, Qing Wang, Zhuoran Yang et al.

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.

Peng Sun, Xinghai Sun, Lei Han et al.

Starcraft II (SC2) is widely considered as the most challenging Real Time Strategy (RTS) game. The underlying challenges include a large observation space, a huge (continuous and infinite) action space, partial observations, simultaneous move for all players, and long horizon delayed rewards for local decisions. To push the frontier of AI research, Deepmind and Blizzard jointly developed the StarCraft II Learning Environment (SC2LE) as a testbench of complex decision making systems. SC2LE provides a few mini games such as MoveToBeacon, CollectMineralShards, and DefeatRoaches, where some AI agents have achieved the performance level of human professional players. However, for full games, the current AI agents are still far from achieving human professional level performance. To bridge this gap, we present two full game AI agents in this paper - the AI agent TStarBot1 is based on deep reinforcement learning over a flat action structure, and the AI agent TStarBot2 is based on hard-coded rules over a hierarchical action structure. Both TStarBot1 and TStarBot2 are able to defeat the built-in AI agents from level 1 to level 10 in a full game (1v1 Zerg-vs-Zerg game on the AbyssalReef map), noting that level 8, level 9, and level 10 are cheating agents with unfair advantages such as full vision on the whole map and resource harvest boosting. To the best of our knowledge, this is the first public work to investigate AI agents that can defeat the built-in AI in the StarCraft II full game.

Qianqian Xu, Jiechao Xiong, Xinwei Sun et al.

A preference order or ranking aggregated from pairwise comparison data is commonly understood as a strict total order. However, in real-world scenarios, some items are intrinsically ambiguous in comparisons, which may very well be an inherent uncertainty of the data. In this case, the conventional total order ranking can not capture such uncertainty with mere global ranking or utility scores. In this paper, we are specifically interested in the recent surge in crowdsourcing applications to predict partial but more accurate (i.e., making less incorrect statements) orders rather than complete ones. To do so, we propose a novel framework to learn some probabilistic models of partial orders as a \emph{margin-based Maximum Likelihood Estimate} (MLE) method. We prove that the induced MLE is a joint convex optimization problem with respect to all the parameters, including the global ranking scores and margin parameter. Moreover, three kinds of generalized linear models are studied, including the basic uniform model, Bradley-Terry model, and Thurstone-Mosteller model, equipped with some theoretical analysis on FDR and Power control for the proposed methods. The validity of these models are supported by experiments with both simulated and real-world datasets, which shows that the proposed models exhibit improvements compared with traditional state-of-the-art algorithms.

Qianqian Xu, Jiechao Xiong, Xiaochun Cao et al.

In crowdsourced preference aggregation, it is often assumed that all the annotators are subject to a common preference or social utility function which generates their comparison behaviors in experiments. However, in reality annotators are subject to variations due to multi-criteria, abnormal, or a mixture of such behaviors. In this paper, we propose a parsimonious mixed-effects model, which takes into account both the fixed effect that the majority of annotators follows a common linear utility model, and the random effect that some annotators might deviate from the common significantly and exhibit strongly personalized preferences. The key algorithm in this paper establishes a dynamic path from the social utility to individual variations, with different levels of sparsity on personalization. The algorithm is based on the Linearized Bregman Iterations, which leads to easy parallel implementations to meet the need of large-scale data analysis. In this unified framework, three kinds of random utility models are presented, including the basic linear model with L2 loss, Bradley-Terry model, and Thurstone-Mosteller model. The validity of these multi-level models are supported by experiments with both simulated and real-world datasets, which shows that the parsimonious multi-level models exhibit improvements in both interpretability and predictive precision compared with traditional HodgeRank.

Ke Ma, Jinshan Zeng, Jiechao Xiong et al.

Learning representation from relative similarity comparisons, often called ordinal embedding, gains rising attention in recent years. Most of the existing methods are batch methods designed mainly based on the convex optimization, say, the projected gradient descent method. However, they are generally time-consuming due to that the singular value decomposition (SVD) is commonly adopted during the update, especially when the data size is very large. To overcome this challenge, we propose a stochastic algorithm called SVRG-SBB, which has the following features: (a) SVD-free via dropping convexity, with good scalability by the use of stochastic algorithm, i.e., stochastic variance reduced gradient (SVRG), and (b) adaptive step size choice via introducing a new stabilized Barzilai-Borwein (SBB) method as the original version for convex problems might fail for the considered stochastic \textit{non-convex} optimization problem. Moreover, we show that the proposed algorithm converges to a stationary point at a rate $\mathcal{O}(\frac{1}{T})$ in our setting, where $T$ is the number of total iterations. Numerous simulations and real-world data experiments are conducted to show the effectiveness of the proposed algorithm via comparing with the state-of-the-art methods, particularly, much lower computational cost with good prediction performance.

Qianqian Xu, Jiechao Xiong, Xi Chen et al.

Recently, crowdsourcing has emerged as an effective paradigm for human-powered large scale problem solving in various domains. However, task requester usually has a limited amount of budget, thus it is desirable to have a policy to wisely allocate the budget to achieve better quality. In this paper, we study the principle of information maximization for active sampling strategies in the framework of HodgeRank, an approach based on Hodge Decomposition of pairwise ranking data with multiple workers. The principle exhibits two scenarios of active sampling: Fisher information maximization that leads to unsupervised sampling based on a sequential maximization of graph algebraic connectivity without considering labels; and Bayesian information maximization that selects samples with the largest information gain from prior to posterior, which gives a supervised sampling involving the labels collected. Experiments show that the proposed methods boost the sampling efficiency as compared to traditional sampling schemes and are thus valuable to practical crowdsourcing experiments.

Qianqian Xu, Ming Yan, Chendi Huang et al.

Outlier detection is a crucial part of robust evaluation for crowdsourceable assessment of Quality of Experience (QoE) and has attracted much attention in recent years. In this paper, we propose some simple and fast algorithms for outlier detection and robust QoE evaluation based on the nonconvex optimization principle. Several iterative procedures are designed with or without knowing the number of outliers in samples. Theoretical analysis is given to show that such procedures can reach statistically good estimates under mild conditions. Finally, experimental results with simulated and real-world crowdsourcing datasets show that the proposed algorithms could produce similar performance to Huber-LASSO approach in robust ranking, yet with nearly 8 or 90 times speed-up, without or with a prior knowledge on the sparsity size of outliers, respectively. Therefore the proposed methodology provides us a set of helpful tools for robust QoE evaluation with crowdsourcing data.

Chendi Huang, Xinwei Sun, Jiechao Xiong et al.

Boosting as gradient descent algorithms is one popular method in machine learning. In this paper a novel Boosting-type algorithm is proposed based on restricted gradient descent with structural sparsity control whose underlying dynamics are governed by differential inclusions. In particular, we present an iterative regularization path with structural sparsity where the parameter is sparse under some linear transforms, based on variable splitting and the Linearized Bregman Iteration. Hence it is called \emph{Split LBI}. Despite its simplicity, Split LBI outperforms the popular generalized Lasso in both theory and experiments. A theory of path consistency is presented that equipped with a proper early stopping, Split LBI may achieve model selection consistency under a family of Irrepresentable Conditions which can be weaker than the necessary and sufficient condition for generalized Lasso. Furthermore, some $\ell_2$ error bounds are also given at the minimax optimal rates. The utility and benefit of the algorithm are illustrated by several applications including image denoising, partial order ranking of sport teams, and world university grouping with crowdsourced ranking data.

Qianqian Xu, Jiechao Xiong, Xiaochun Cao et al.

In crowdsourced preference aggregation, it is often assumed that all the annotators are subject to a common preference or utility function which generates their comparison behaviors in experiments. However, in reality annotators are subject to variations due to multi-criteria, abnormal, or a mixture of such behaviors. In this paper, we propose a parsimonious mixed-effects model based on HodgeRank, which takes into account both the fixed effect that the majority of annotators follows a common linear utility model, and the random effect that a small subset of annotators might deviate from the common significantly and exhibits strongly personalized preferences. HodgeRank has been successfully applied to subjective quality evaluation of multimedia and resolves pairwise crowdsourced ranking data into a global consensus ranking and cyclic conflicts of interests. As an extension, our proposed methodology further explores the conflicts of interests through the random effect in annotator specific variations. The key algorithm in this paper establishes a dynamic path from the common utility to individual variations, with different levels of parsimony or sparsity on personalization, based on newly developed Linearized Bregman Algorithms with Inverse Scale Space method. Finally the validity of the methodology are supported by experiments with both simulated examples and three real-world crowdsourcing datasets, which shows that our proposed method exhibits better performance (i.e. smaller test error) compared with HodgeRank due to its parsimonious property.

Qianqian Xu, Jiechao Xiong, Xiaochun Cao et al.

With the rapid growth of crowdsourcing platforms it has become easy and relatively inexpensive to collect a dataset labeled by multiple annotators in a short time. However due to the lack of control over the quality of the annotators, some abnormal annotators may be affected by position bias which can potentially degrade the quality of the final consensus labels. In this paper we introduce a statistical framework to model and detect annotator's position bias in order to control the false discovery rate (FDR) without a prior knowledge on the amount of biased annotators - the expected fraction of false discoveries among all discoveries being not too high, in order to assure that most of the discoveries are indeed true and replicable. The key technical development relies on some new knockoff filters adapted to our problem and new algorithms based on the Inverse Scale Space dynamics whose discretization is potentially suitable for large scale crowdsourcing data analysis. Our studies are supported by experiments with both simulated examples and real-world data. The proposed framework provides us a useful tool for quantitatively studying annotator's abnormal behavior in crowdsourcing data arising from machine learning, sociology, computer vision, multimedia, etc.

Braxton Osting, Jiechao Xiong, Qianqian Xu et al.

Crowdsourcing platforms are now extensively used for conducting subjective pairwise comparison studies. In this setting, a pairwise comparison dataset is typically gathered via random sampling, either \emph{with} or \emph{without} replacement. In this paper, we use tools from random graph theory to analyze these two random sampling methods for the HodgeRank estimator. Using the Fiedler value of the graph as a measurement for estimator stability (informativeness), we provide a new estimate of the Fiedler value for these two random graph models. In the asymptotic limit as the number of vertices tends to infinity, we prove the validity of the estimate. Based on our findings, for a small number of items to be compared, we recommend a two-stage sampling strategy where a greedy sampling method is used initially and random sampling \emph{without} replacement is used in the second stage. When a large number of items is to be compared, we recommend random sampling with replacement as this is computationally inexpensive and trivially parallelizable. Experiments on synthetic and real-world datasets support our analysis.

Yanwei Fu, Timothy M. Hospedales, Tao Xiang et al.

The problem of estimating subjective visual properties from image and video has attracted increasing interest. A subjective visual property is useful either on its own (e.g. image and video interestingness) or as an intermediate representation for visual recognition (e.g. a relative attribute). Due to its ambiguous nature, annotating the value of a subjective visual property for learning a prediction model is challenging. To make the annotation more reliable, recent studies employ crowdsourcing tools to collect pairwise comparison labels because human annotators are much better at ranking two images/videos (e.g. which one is more interesting) than giving an absolute value to each of them separately. However, using crowdsourced data also introduces outliers. Existing methods rely on majority voting to prune the annotation outliers/errors. They thus require large amount of pairwise labels to be collected. More importantly as a local outlier detection method, majority voting is ineffective in identifying outliers that can cause global ranking inconsistencies. In this paper, we propose a more principled way to identify annotation outliers by formulating the subjective visual property prediction task as a unified robust learning to rank problem, tackling both the outlier detection and learning to rank jointly. Differing from existing methods, the proposed method integrates local pairwise comparison labels together to minimise a cost that corresponds to global inconsistency of ranking order. This not only leads to better detection of annotation outliers but also enables learning with extremely sparse annotations. Extensive experiments on various benchmark datasets demonstrate that our new approach significantly outperforms state-of-the-arts alternatives.

Qianqian Xu, Jiechao Xiong, Xiaochun Cao et al.

Nowadays, how to effectively evaluate visual properties has become a popular topic for fine-grained visual comprehension. In this paper we study the problem of how to estimate such visual properties from a ranking perspective with the help of the annotators from online crowdsourcing platforms. The main challenges of our task are two-fold. On one hand, the annotations often contain contaminated information, where a small fraction of label flips might ruin the global ranking of the whole dataset. On the other hand, considering the large data capacity, the annotations are often far from being complete. What is worse, there might even exist imbalanced annotations where a small subset of samples are frequently annotated. Facing such challenges, we propose a robust ranking framework based on the principle of Hodge decomposition of imbalanced and incomplete ranking data. According to the HodgeRank theory, we find that the major source of the contamination comes from the cyclic ranking component of the Hodge decomposition. This leads us to an outlier detection formulation as sparse approximations of the cyclic ranking projection. Taking a step further, it facilitates a novel outlier detection model as Huber's LASSO in robust statistics. Moreover, simple yet scalable algorithms are developed based on Linearized Bregman Iteration to achieve an even less biased estimator. Statistical consistency of outlier detection is established in both cases under nearly the same conditions. Our studies are supported by experiments with both simulated examples and real-world data. The proposed framework provides us a promising tool for robust ranking with large scale crowdsourcing data arising from computer vision.

Stanley Osher, Feng Ruan, Jiechao Xiong et al.

In this paper, we recover sparse signals from their noisy linear measurements by solving nonlinear differential inclusions, which is based on the notion of inverse scale space (ISS) developed in applied mathematics. Our goal here is to bring this idea to address a challenging problem in statistics, \emph{i.e.} finding the oracle estimator which is unbiased and sign-consistent using dynamics. We call our dynamics \emph{Bregman ISS} and \emph{Linearized Bregman ISS}. A well-known shortcoming of LASSO and any convex regularization approaches lies in the bias of estimators. However, we show that under proper conditions, there exists a bias-free and sign-consistent point on the solution paths of such dynamics, which corresponds to a signal that is the unbiased estimate of the true signal and whose entries have the same signs as those of the true signs, \emph{i.e.} the oracle estimator. Therefore, their solution paths are regularization paths better than the LASSO regularization path, since the points on the latter path are biased when sign-consistency is reached. We also show how to efficiently compute their solution paths in both continuous and discretized settings: the full solution paths can be exactly computed piece by piece, and a discretization leads to \emph{Linearized Bregman iteration}, which is a simple iterative thresholding rule and easy to parallelize. Theoretical guarantees such as sign-consistency and minimax optimal $l_2$-error bounds are established in both continuous and discrete settings for specific points on the paths. Early-stopping rules for identifying these points are given. The key treatment relies on the development of differential inequalities for differential inclusions and their discretizations, which extends the previous results and leads to exponentially fast recovering of sparse signals before selecting wrong ones.