Nian Ji

Xuanji Xiao, Huabin Chen, Yuzhen Liu et al.

Click-through rate (CTR) and post-click conversion rate (CVR) predictions are two fundamental modules in industrial ranking systems such as recommender systems, advertising, and search engines. Since CVR involves much fewer samples than CTR (known as the CVR data sparsity problem), most of the existing works try to leverage CTR&CVR multi-task learning to improve CVR performance. However, typical coarse-grained sub-network/layer sharing methods may introduce conflicts and lead to performance degradation, since not every neuron or neuron connection in one layer should be shared between CVR and CTR tasks. This is because users may have different fine-grained content feature preferences between deep consumption and click behavior, represented by CVR and CTR, respectively. To address this sharing&conflict problem, we propose a novel multi-task CVR modeling scheme with neuron-connection level sharing named NCS4CVR, which can automatically and flexibly learn which neuron weights are shared or not shared without artificial experience. Compared with previous layer-level sharing methods, this is the first time that a fine-grained CTR&CVR sharing method at the neuron connection level is proposed, which is a research paradigm shift in the sharing level. Both offline and online experiments demonstrate that our method outperforms both the single-task model and the layer-level sharing model. Our proposed method has now been successfully deployed in an industry video recommender system serving major traffic.

Paul Ratazzi, Ashok Bommisetti, Nian Ji et al.

Virtualization is frequently used to isolate untrusted processes and control their access to sensitive resources. However, isolation usually carries a price in terms of less resource sharing and reduced inter-process communication. In an open architecture such as Android, this price and its impact on performance, usability, and transparency must be carefully considered. Although previous efforts in developing general-purpose isolation solutions have shown that some of these negative side effects can be mitigated, doing so involves overcoming significant design challenges by incorporating numerous additional platform complexities not directly related to improved security. Thus, the general purpose solutions become inefficient and burdensome if the end-user has only specific security goals. In this paper, we present PINPOINT, a resource isolation strategy that forgoes general-purpose solutions in favor of a "building block" approach that addresses specific end-user security goals. PINPOINT embodies the concept of Linux Namespace lightweight isolation, but does so in the Android Framework by guiding the security designer towards isolation points that are contextually close to the resource(s) that need to be isolated. This strategy allows the rest of the Framework to function fully as intended, transparently. We demonstrate our strategy with a case study on Android System Services, and show four applications of PINPOINTed system services functioning with unmodified market apps. Our evaluation results show that practical security and privacy advantages can be gained using our approach, without inducing the problematic side-effects that other general-purpose designs must address.