Visual media have important persuasive power, but prior computer vision approaches have predominantly ignored the persuasive aspects of images. In this work, we propose a suite of data and techniques that enable progress on understanding the messages that visual advertisements convey. We make available a dataset of 64,832 image ads and 3,477 video ads, annotated with ten types of information: the topic and sentiment of the ad; whether it is funny, exciting, or effective; what action it prompts the viewer to do, and what arguments it provides for why this action should be taken; symbolic associations that the ad relies on; the metaphorical object transformations on which especially creative ads rely; and the climax in video ads. We develop methods that use multimodal cues, i.e., both visuals and slogans, for both the image and video domains. Our methods rely on finding poignant content spatially and temporally. We also examine the creative story construction in ads: for videos, we learn to predict when the climax occurs (if any), and how effective the story is; for images, we analyze how object transformations in ads metaphorically depict product properties.

Reliability-aware speedup models study the expected speedup of a parallel application as a function of the number of processors, on a platform susceptible to processor failures. Existing works in this area have developed models using checkpoint-restart (without replication) as the only fault tolerance mechanism, and have studied the upper bound on the number of processors beyond which the application speedup starts to degrade due to increasing likelihood of failure. In this work, we develop speedup models in which replication, specifically dual replication, is also employed for resilience. We demonstrate that the upper bound on the number of processors to execute a perfectly parallel application using dual replication is of the order λ^-^2 where λ is the individual processor failure rate. We also compare the dual replication model with that of no-replication. Specifically, we found that, given the same hardware resources, replication starts offering better speedup just before the upper bound on the number of processors for no-replication is reached. Taken together, our results indicate that replication can significantly enhance reliability-aware speedup models by i) pushing the number of processors that yield the optimal speedup to a much higher value than what is possible without replication, and ii) improving on the optimal speedup possible through checkpoint-restart alone.

In-memory checkpointing has increased in popularity over the years because it significantly improves the time to take a checkpoint. It is usually accomplished by placing all or part of a processor's checkpoint into the local memory of a remote node within the cluster. If, however, the checkpointed node and the node containing its checkpoint both fail in quick succession, recovery using in-memory checkpoints becomes impossible. In this paper, we explore the problem of placing in-memory checkpoints among nodes whose individual failure likelihoods are not identical. We provide theoretical results on the optimal way to place in-memory checkpoints such that the probability of occurrence of a catastrophic failure, i.e. failure of a node as well as the node containing its checkpoint, is minimized. Using the failure logs spread over 5 years of a 49,152 node supercomputer, we show that checkpoint placement schemes that utilize knowledge of node failure likelihoods, and are guided by the theoretical results we provide, can significantly reduce the total number of such catastrophic failures when compared with placement schemes that are oblivious of the heterogeneity in nodes based on their failure likelihoods.

With the increase in scale of HPC systems, the frequency of system wide failures is expected to increase. The performance of Coordinated Checkpoint/Restart (C/R), the traditional fault tolerance technique, degrades under high failure rates because of frequent global rollbacks, which themselves are susceptible to failures. We propose CoLoR, a fault tolerance scheme that i)requires only the failing process to recover, ii)overlaps reexecution with restart, and iii)avoids the cumulative effect of successive failures. Our theoretical analysis reveals that such a scheme results in lower expected completion time than coordinated C/R. We also provide a proof-of-concept implementation in MPI using receiver based message logging and colocated rescuer (CoLoR)processes, and evaluate its performance on several HPC benchmarks. Our experimental results, combined with observations from the theoretical analysis, show that CoLoR can outperform both traditional C/R and replication over a large range of system sizes, without using extra logger nodes.

There is more to images than their objective physical content: for example, advertisements are created to persuade a viewer to take a certain action. We propose the novel problem of automatic advertisement understanding. To enable research on this problem, we create two datasets: an image dataset of 64,832 image ads, and a video dataset of 3,477 ads. Our data contains rich annotations encompassing the topic and sentiment of the ads, questions and answers describing what actions the viewer is prompted to take and the reasoning that the ad presents to persuade the viewer (What should I do according to this ad, and why should I do it?), and symbolic references ads make (e.g. a dove symbolizes peace). We also analyze the most common persuasive strategies ads use, and the capabilities that computer vision systems should have to understand these strategies. We present baseline classification results for several prediction tasks, including automatically answering questions about the messages of the ads.