We are in the dawn of deep learning explosion for smartphones. To bridge the gap between research and practice, we present the first empirical study on 16,500 the most popular Android apps, demystifying how smartphone apps exploit deep learning in the wild. To this end, we build a new static tool that dissects apps and analyzes their deep learning functions. Our study answers threefold questions: what are the early adopter apps of deep learning, what do they use deep learning for, and how do their deep learning models look like. Our study has strong implications for app developers, smartphone vendors, and deep learning R&D. On one hand, our findings paint a promising picture of deep learning for smartphones, showing the prosperity of mobile deep learning frameworks as well as the prosperity of apps building their cores atop deep learning. On the other hand, our findings urge optimizations on deep learning models deployed on smartphones, protection of these models, and validation of research ideas on these models.

No abstract

Deep learning (DL) becomes increasingly pervasive, being used in a wide range of software applications. These software applications, named as DL based software (in short as DL software), integrate DL models trained using a large data corpus with DL programs written based on DL frameworks such as TensorFlow and Keras. A DL program encodes the network structure of a desirable DL model and the process by which the model is trained using the training data. To help developers of DL software meet the new challenges posed by DL, enormous research e orts in software engineering have been devoted. Existing studies focus on the development of DL software and extensively analyze faults in DL programs. However, the deployment of DL software has not been comprehensively studied. To ll this knowledge gap, this paper presents a comprehensive study on understanding challenges in deploying DL software. We mine and analyze 3,023 relevant posts from Stack Over ow, a popular Q&A website for developers, and show the increasing popularity and high di culty of DL software deployment among developers. We build a taxonomy of speci c challenges encountered by developers in the process of DL software deployment through manual inspection of 769 sampled posts and report a series of actionable implications for researchers, developers, and DL framework vendors. CCS CONCEPTS • Software and its engineering → Software creation and management; • Computing methodologies → Arti cial intelligence; • General and reference → Empirical studies.

No abstract

Antiadhesion barriers such as films and hydrogels used to wrap repaired tendons are important for preventing the formation of adhesion tissue after tendon surgery. However, sliding of the tendon can compress the adjacent hydrogel barrier and cause it to rupture, which may then lead to unexpected inflammation. Here, a self‐healing and deformable hyaluronic acid (HA) hydrogel is constructed as a peritendinous antiadhesion barrier. Matrix metalloproteinase‐2 (MMP‐2)‐degradable gelatin‐methacryloyl (GelMA) microspheres (MSs) encapsulated with Smad3‐siRNA nanoparticles are entrapped within the HA hydrogel to inhibit fibroblast proliferation and prevent peritendinous adhesion. GelMA MSs are responsively degraded by upregulation of MMP‐2, achieving on‐demand release of siRNA nanoparticles. Silencing effect of Smad3‐siRNA nanoparticles is around 75% toward targeted gene. Furthermore, the self‐healing hydrogel shows relatively attenuated inflammation compared to non‐healing hydrogel. The mean adhesion scores of composite barrier group are 1.67 ± 0.51 and 2.17 ± 0.75 by macroscopic and histological evaluation, respectively. The proposed self‐healing hydrogel antiadhesion barrier with MMP‐2‐responsive drug release behavior is highly effective for decreasing inflammation and inhibiting tendon adhesion. Therefore, this research provides a new strategy for the development of safe and effective antiadhesion barriers.