Siyu Gao

Geng Chen, Wendong Zhang, Han Lu et al.

Predictive learning ideally builds the world model of physical processes in one or more given environments. Typical setups assume that we can collect data from all environments at all times. In practice, however, different prediction tasks may arrive sequentially so that the environments may change persistently throughout the training procedure. Can we develop predictive learning algorithms that can deal with more realistic, non-stationary physical environments? In this paper, we study a new continual learning problem in the context of video prediction, and observe that most existing methods suffer from severe catastrophic forgetting in this setup. To tackle this problem, we propose the continual predictive learning (CPL) approach, which learns a mixture world model via predictive experience replay and performs test-time adaptation with non-parametric task inference. We construct two new benchmarks based on RoboNet and KTH, in which different tasks correspond to different physical robotic environments or human actions. Our approach is shown to effectively mitigate forgetting and remarkably outperform the naïve combinations of previous art in video prediction and continual learning.

Yanpeng Zhao, Yiwei Hao, Siyu Gao et al.

Learning object-centric representations from unsupervised videos is challenging. Unlike most previous approaches that focus on decomposing 2D images, we present a 3D generative model named DynaVol-S for dynamic scenes that enables object-centric learning within a differentiable volume rendering framework. The key idea is to perform object-centric voxelization to capture the 3D nature of the scene, which infers per-object occupancy probabilities at individual spatial locations. These voxel features evolve through a canonical-space deformation function and are optimized in an inverse rendering pipeline with a compositional NeRF. Additionally, our approach integrates 2D semantic features to create 3D semantic grids, representing the scene through multiple disentangled voxel grids. DynaVol-S significantly outperforms existing models in both novel view synthesis and unsupervised decomposition tasks for dynamic scenes. By jointly considering geometric structures and semantic features, it effectively addresses challenging real-world scenarios involving complex object interactions. Furthermore, once trained, the explicitly meaningful voxel features enable additional capabilities that 2D scene decomposition methods cannot achieve, such as novel scene generation through editing geometric shapes or manipulating the motion trajectories of objects.

Minting Pan, Wendong Zhang, Geng Chen et al.

Learning physical dynamics in a series of non-stationary environments is a challenging but essential task for model-based reinforcement learning (MBRL) with visual inputs. It requires the agent to consistently adapt to novel tasks without forgetting previous knowledge. In this paper, we present a new continual learning approach for visual dynamics modeling and explore its efficacy in visual control. The key assumption is that an ideal world model can provide a non-forgetting environment simulator, which enables the agent to optimize the policy in a multi-task learning manner based on the imagined trajectories from the world model. To this end, we first introduce the life-long world model, which learns task-specific latent dynamics using a mixture of Gaussians and incorporates generative experience replay to mitigate catastrophic forgetting. Then, we further address the value estimation challenge for previous tasks with the exploratory-conservative behavior learning approach. Our model remarkably outperforms the straightforward combinations of existing continual learning and visual RL algorithms on DeepMind Control Suite and Meta-World benchmarks with continual visual control tasks.

Yanpeng Zhao, Siyu Gao, Yunbo Wang et al.

Unsupervised learning of object-centric representations in dynamic visual scenes is challenging. Unlike most previous approaches that learn to decompose 2D images, we present DynaVol, a 3D scene generative model that unifies geometric structures and object-centric learning in a differentiable volume rendering framework. The key idea is to perform object-centric voxelization to capture the 3D nature of the scene, which infers the probability distribution over objects at individual spatial locations. These voxel features evolve over time through a canonical-space deformation function, forming the basis for global representation learning via slot attention. The voxel features and global features are complementary and are both leveraged by a compositional NeRF decoder for volume rendering. DynaVol remarkably outperforms existing approaches for unsupervised dynamic scene decomposition. Once trained, the explicitly meaningful voxel features enable additional capabilities that 2D scene decomposition methods cannot achieve: it is possible to freely edit the geometric shapes or manipulate the motion trajectories of the objects.

Haochen Yuan, Minting Pan, Yunbo Wang et al.

Reinforcement learning (RL) has shown significant promise for sequential portfolio optimization tasks, such as stock trading, where the objective is to maximize cumulative returns while minimizing risks using historical data. However, traditional RL approaches often produce policies that merely memorize the optimal yet impractical buying and selling behaviors within the fixed dataset. These offline policies are less generalizable as they fail to account for the non-stationary nature of the market. Our approach, MetaTrader, frames portfolio optimization as a new type of partial-offline RL problem and makes two technical contributions. First, MetaTrader employs a bilevel learning framework that explicitly trains the RL agent to improve both in-domain profits on the original dataset and out-of-domain performance across diverse transformations of the raw financial data. Second, our approach incorporates a new temporal difference (TD) method that approximates worst-case TD estimates from a batch of transformed TD targets, addressing the value overestimation issue that is particularly challenging in scenarios with limited offline data. Our empirical results on two public stock datasets show that MetaTrader outperforms existing methods, including both RL-based approaches and traditional stock prediction models.

Jungang Zou, Fan Lin, Siyu Gao et al.

Dynamic community detection is the hotspot and basic problem of complex network and artificial intelligence research in recent years. It is necessary to maximize the accuracy of clustering as the network structure changes, but also to minimize the two consecutive clustering differences between the two results. There is a trade-off relationship between these two objectives. In this paper, we propose a Feature Transfer Based Multi-Objective Optimization Genetic Algorithm (TMOGA) based on transfer learning and traditional multi-objective evolutionary algorithm framework. The main idea is to extract stable features from past community structures, retain valuable feature information, and integrate this feature information into current optimization processes to improve the evolutionary algorithms. Additionally, a new theoretical framework is proposed in this paper to analyze community detection problem based on information theory. Then, we exploit this framework to prove the rationality of TMOGA. Finally, the experimental results show that our algorithm can achieve better clustering effects compared with the state-of-the-art dynamic network community detection algorithms in diverse test problems.

Zheyuan Ryan Shi, Aaron Schlenker, Brian Hay et al.

Cyber adversaries have increasingly leveraged social engineering attacks to breach large organizations and threaten the well-being of today's online users. One clever technique, the "watering hole" attack, compromises a legitimate website to execute drive-by download attacks by redirecting users to another malicious domain. We introduce a game-theoretic model that captures the salient aspects for an organization protecting itself from a watering hole attack by altering the environment information in web traffic so as to deceive the attackers. Our main contributions are (1) a novel Social Engineering Deception (SED) game model that features a continuous action set for the attacker, (2) an in-depth analysis of the SED model to identify computationally feasible real-world cases, and (3) the CyberTWEAK algorithm which solves for the optimal protection policy. To illustrate the potential use of our framework, we built a browser extension based on our algorithms which is now publicly available online. The CyberTWEAK extension will be vital to the continued development and deployment of countermeasures for social engineering.