We would like to sincerely thank all our reviewers for their valuable feedback and insightful comments. In experiments on generalizability, we've used Appendix A.3 describes how we generated neural networks for each of these tasks using the ENAS We will clarify this point in the revision. We've chosen REINFORCE for RL opti-12 We leave this investigation for future work. Placeto's policy is computed by running two separate, indepen-18 We will revise Section 2.2 and Figure 3 to clarify this point. We will add these details to the appendix section A.7 and additionally, release a source file with all of of our We will revise Section 2.1 to Stopping criterion used for Section 4.1 (Review #2): Thank you for the suggestion!

Resettlement agencies have started to adopt data-driven algorithmic matching to match refugees to locations using employment rate as a measure of utility. Given a pool of refugees, data-driven algorithmic matching utilizes a classifier to predict the probability that each refugee would find employment at any given location. Then, it uses the predicted probabilities to estimate the expected utility of all possible placement decisions. Finally, it finds the placement decisions that maximize the predicted utility by solving a maximum weight bipartite matching problem. In this work, we argue that, using existing solutions, there may be pools of refugees for which data-driven algorithmic matching is (counterfactually) harmful -- it would have achieved lower utility than a given default policy used in the past, had it been used. Then, we develop a post-processing algorithm that, given placement decisions made by a default policy on a pool of refugees and their employment outcomes, solves an inverse~matching problem to minimally modify the predictions made by a given classifier. Under these modified predictions, the optimal matching policy that maximizes predicted utility on the pool is guaranteed to be not harmful. Further, we introduce a Transformer model that, given placement decisions made by a default policy on multiple pools of refugees and their employment outcomes, learns to modify the predictions made by a classifier so that the optimal matching policy that maximizes predicted utility under the modified predictions on an unseen pool of refugees is less likely to be harmful than under the original predictions. Experiments on simulated resettlement processes using synthetic refugee data created from a variety of publicly available data suggest that our methodology may be effective in making algorithmic placement decisions that are less likely to be harmful than existing solutions.

Hybrid storage systems (HSS) use multiple different storage devices to provide high and scalable storage capacity at high performance. Recent research proposes various techniques that aim to accurately identify performance-critical data to place it in a "best-fit" storage device. Unfortunately, most of these techniques are rigid, which (1) limits their adaptivity to perform well for a wide range of workloads and storage device configurations, and (2) makes it difficult for designers to extend these techniques to different storage system configurations (e.g., with a different number or different types of storage devices) than the configuration they are designed for. We introduce Sibyl, the first technique that uses reinforcement learning for data placement in hybrid storage systems. Sibyl observes different features of the running workload as well as the storage devices to make system-aware data placement decisions. For every decision it makes, Sibyl receives a reward from the system that it uses to evaluate the long-term performance impact of its decision and continuously optimizes its data placement policy online. We implement Sibyl on real systems with various HSS configurations. Our results show that Sibyl provides 21.6%/19.9% performance improvement in a performance-oriented/cost-oriented HSS configuration compared to the best previous data placement technique. Our evaluation using an HSS configuration with three different storage devices shows that Sibyl outperforms the state-of-the-art data placement policy by 23.9%-48.2%, while significantly reducing the system architect's burden in designing a data placement mechanism that can simultaneously incorporate three storage devices. We show that Sibyl achieves 80% of the performance of an oracle policy that has complete knowledge of future access patterns while incurring a very modest storage overhead of only 124.4 KiB.

Sociologists of education increasingly highlight the role of opportunity hoarding in the formation of Black-White educational inequalities. Informed by this literature, this article unpacks the necessary and sufficient conditions under which the hoarding of educational resources emerges within schools. It develops a qualitatively informed agent-based model which captures Black and White students' competition for a valuable school resource: advanced coursework. In contrast to traditional accounts -- which explain the emergence of hoarding through the actions of Whites that keep valuable resources within White communities -- simulations, perhaps surprisingly, show hoarding to arise even when Whites do not play the role of hoarders of resources. Behind this result is the fact that a structural inequality (i.e., racial differences in social class) -- and not action-driven hoarding -- is the necessary condition for hoarding to emerge. Findings, therefore, illustrate that common action-driven understandings of opportunity hoarding can overlook the structural foundations behind this important phenomenon. Policy implications are discussed.

We propose to demonstrate a network slice placement optimization solution based on Deep Reinforcement Learning (DRL), referred to as Heuristically-controlled DRL, which uses a heuristic to control the DRL algorithm convergence. The solution is adapted to realistic networks with large scale and under non-stationary traffic conditions (namely, the network load). We demonstrate the applicability of the proposed solution and its higher and stable performance over a non-controlled DRL-based solution. Demonstration scenarios include full online learning with multiple volatile network slice placement request arrivals.

We've all had noisy neighbors at one point in our life. Whether it's at a cafe or through a wall of an apartment, it is always disruptive. The need for good manners in shared spaces turns out to be important not just for people, but for your Docker containers too. When you're running in the cloud your containers are in a shared space; in particular they share the CPU's memory hierarchy of the host instance. Because microprocessors are so fast, computer architecture design has evolved towards adding various levels of caching between compute units and the main memory, in order to hide the latency of bringing the bits to the brains.