Pipelined SGD (Pipe-SGD)
Distributed training of deep nets is an important technique to address some of the present day computing challenges like memory consumption and computational demands. Classical distributed approaches, synchronous or asynchronous, are based on the parameter server architecture, i.e., worker nodes compute gradients which are communicated to the parameter server while updated parameters are returned. Recently, distributed training with AllReduce operations gained popularity as well. While many of those operations seem appealing, little is reported about wall-clock training time improvements. In this paper, we carefully analyze the AllReduce based setup, propose timing models which include network latency, bandwidth, cluster size and compute time, and demonstrate that a pipelined training with a width of two combines the best of both synchronous and asynchronous training. Specifically, for a setup consisting of a four-node GPU cluster we show wall-clock time training improvements of up to 5.4x compared to conventional approaches. …

Differential Fairness
We introduce a measure of fairness for algorithms and data with regard to multiple protected attributes. Our proposed definition, differential fairness, is informed by the framework of intersectionality, which analyzes how interlocking systems of power and oppression affect individuals along overlapping dimensions including race, gender, sexual orientation, class, and disability. We show that our criterion behaves sensibly for any subset of the protected attributes, and we illustrate links to differential privacy. A case study on census data demonstrates the utility of our approach. …

Shallow-Deep Network (SDN)
While deep neural networks (DNNs) can perform complex classification tasks, most of their natural inputs do not necessitate the depth of the modern architectures. This leads to wasted computation, as the network overthinks on the simpler inputs. The overthinking problem could be prevented if standard DNNs could produce early predictions. However, prior work suggests that this is challenging in existing architectures, such as ResNet, as their internal layers are not trained for classification and optimizing them for accurate predictions hurts the end performance. In this paper, we explore the overthinking problem, and, as a remedy, we propose a generic modification to off-the-shelf DNNs—the Shallow-Deep Network (SDN). With this modification, a DNN can efficiently produce predictions from either shallow or deep layers, as appropriate for the given input. We employ feature reduction and a layer-wise objective function to train these progressively deeper internal classifiers while preserving the end-performance. We can apply the SDN modification either by training from scratch or by tuning a pre-trained model. Experiments on four architectures (VGG, ResNet, WideResNet, and MobileNet) and three image classifications tasks suggest that, for an average input, an SDN can produce a correct prediction before its middle layer. By avoiding unnecessary computation, the SDN can reduce the required number of operations for an input by 41% over the original network. Finally, we observe that disagreements among the early classifiers reliably indicate inputs where the network is likely to make a mistake. Building on this observation we propose an internal confusion metric and a method to diagnose misclassifications by visualizing these disagreements. …

Sharpness
It is well-known that, without restricting treatment effect heterogeneity, instrumental variable (IV) methods only identify ‘local’ effects among compliers, i.e., those subjects who take treatment only when encouraged by the IV. Local effects are controversial since they seem to only apply to an unidentified subgroup; this has led many to denounce these effects as having little policy relevance. However, we show that such pessimism is not always warranted: it is possible in some cases to accurately predict who compliers are, and obtain tight bounds on more generalizable effects in identifiable subgroups. We propose methods for doing so and study their estimation error and asymptotic properties, showing that these tasks can in theory be accomplished even with very weak IVs. We go on to introduce a new measure of IV quality called ‘sharpness’, which reflects the variation in compliance explained by covariates, and captures how well one can identify compliers and obtain tight bounds on identifiable subgroup effects. We develop an estimator of sharpness, and show that it is asymptotically efficient under weak conditions. Finally we explore finite-sample properties via simulation, and apply the methods to study canvassing effects on voter turnout. We propose that sharpness should be presented alongside strength to assess IV quality. …