Ryan Wang

The rising complexity of deep neural networks has raised rigorous demands for computational hardware and deployment expertise. As an alternative, outsourcing a pre-trained model to a third party server has been increasingly prevalent. However, it creates opportunities for attackers to interfere with the prediction outcomes of the deep neural network. In this paper, we focus on integrity verification of the prediction results from outsourced deep neural models and make a thread of contributions.… Show more

The rising complexity of deep neural networks has raised rigorous demands for computational hardware and deployment expertise. As an alternative, outsourcing a pre-trained model to a third party server has been increasingly prevalent. However, it creates opportunities for attackers to interfere with the prediction outcomes of the deep neural network. In this paper, we focus on integrity verification of the prediction results from outsourced deep neural models and make a thread of contributions. We propose a new attack based on steganography that enables the server to generate wrong prediction results in a command-and-control fashion. Following that, we design a homomorphic encryption-based authentication scheme to detect wrong predictions made by any attack. Our extensive experiments on benchmark datasets demonstrate the invisibility of the attack and the effectiveness of our authentication approach. Show less

https://bigdataieee.org/BigData2020/AcceptedPapers.html

In the mobile gaming industry, organic installs refer to downloads that cannot be attributed to any advertising channel and thus do not introduce upfront user acquisition (UA) cost. Understanding the causal factors on organic installs is of vital importance for a game’s ecosystem, as such knowledge can help bring in more organic users, who tend to be more loyal and active. A major challenge in discovering the causal effects is the… Show more

https://bigdataieee.org/BigData2020/AcceptedPapers.html

In the mobile gaming industry, organic installs refer to downloads that cannot be attributed to any advertising channel and thus do not introduce upfront user acquisition (UA) cost. Understanding the causal factors on organic installs is of vital importance for a game’s ecosystem, as such knowledge can help bring in more organic users, who tend to be more loyal and active. A major challenge in discovering the causal effects is the potential temporal lag between an UA operation and the growth in organic installs. In this paper, we solve the problem by using a deep attentional neural network to analyze multivariate time series data. The core of our design is a novel attention mechanism, namely 2D-ATT, that can learn the contribution of each feature to the target at different levels of temporal delay. Our experiments on a series of synthetic datasets show that 2D-ATT outperforms existing approaches for discovering complex causal effects. We also use 2D-ATT to analyze a real-world mobile game dataset collected by Jam City, a video game company based in California. Our discoveries provide valuable insights to UA operations. Show less

We present the prototype of an egocentric vision based assistive co-robot system. In this co-robot system, the user is wearing a pair of glasses with a forward looking camera, and is actively engaged in the control loop of the robot in navigational tasks.
The egocentric vision glasses serve for two purposes. First, it serves as a source of visual input to request the robot to find a certain object in the environment. Second, the motion patterns computed from the egocentric video associated… Show more

We present the prototype of an egocentric vision based assistive co-robot system. In this co-robot system, the user is wearing a pair of glasses with a forward looking camera, and is actively engaged in the control loop of the robot in navigational tasks.
The egocentric vision glasses serve for two purposes. First, it serves as a source of visual input to request the robot to find a certain object in the environment. Second, the motion patterns computed from the egocentric video associated with a specific set of head movements are exploited to guide the robot to find the object.
These are especially helpful for quadriplegic individuals who do not have needed hand functionality for interaction and control with other modalities (e.g., joystick). In our co-robot system, when the robot does not fulfill the object finding task in a pre-specified time window, it would actively solicit user controls for guidance. Then the users can use the egocentric vision based gesture interface to orient the robot towards the direction of the object. After that the robot will automatically navigate towards the object until it finds it.
Our experiments validated the efficacy of the closed-loop design to engage the human in the loop. Show less