Search

Lead halide perovskites have attracted considerable attention as potential gas sensing elements due to their unique ability to detect and respond to an external stimulus with measurable electrical or optical signals. The distinctive characteristics of these materials lie in their ability to operate upon gas exposure at room temperature. However, the presence of lead (Pb) poses a serious challenge for their widespread application and commercialization, owing to its toxicity. To address this issue, lead-free Cs2AgBiBr6 double perovskite is explored as a promising alternative sensing element for room temperature gas detection. This sensor can be regarded as eco-friendly for the following reasons: it does not contain Pb; it is synthesized at room temperature without using strong organic solvents; and it operates and recovers at room temperature without heating or UV irradiation and with a very low input voltage (0.1V), thus reducing overall energy consumption. The influence of the perovskite morphology on the ozone (O3) sensing performance is investigated, as well as the sensing stability under varying conditions of time, humidity, and temperature. Notably, the sensors' exceptional selectivity for O3 over other gases and the interaction between the gases and the perovskite surface is confirmed through experimental and first-principles calculations., Comment: Main document: 37 pages, 7 figures; Support Information: 14 pages, 15 figures; Corresponding authors: Brintakis, Kostopoulou, Stratakis

Modern machine learning systems represent their computations as dataflow graphs. The increasingly complex neural network architectures crave for more powerful yet efficient programming abstractions. In this paper we propose an efficient technique for supporting recursive function definitions in dataflow-based systems such as TensorFlow. The proposed approach transforms the given recursive definitions into a static dataflow graph that is enriched with two simple yet powerful dataflow operations. Since static graphs do not change during execution, they can be easily partitioned and executed efficiently in distributed and heterogeneous environments. The proposed technique makes heavy use of the idea of tagging, which was one of the cornerstones of dataflow systems since their inception. We demonstrate that our technique is compatible with the idea of automatic differentiation, a notion that is crucial for dataflow systems that focus on deep learning applications. We describe the principles of an actual implementation of the technique in the TensorFlow framework, and present experimental results that demonstrate that the use of tagging is of paramount importance for developing efficient high-level abstractions for modern dataflow systems.

In recent years, various methods have been introduced for explaining the outputs of "black-box" AI models. However, it is not well understood whether users actually comprehend and trust these explanations. In this paper, we focus on explanations for a regression tool for assessing cancer risk and examine the effect of the explanations' content and format on the user-centric metrics of comprehension and trust. Regarding content, we experiment with two explanation methods: the popular SHAP, based on game-theoretic notions and thus potentially complex for everyday users to comprehend, and occlusion-1, based on feature occlusion which may be more comprehensible. Regarding format, we present SHAP explanations as charts (SC), as is conventional, and occlusion-1 explanations as charts (OC) as well as text (OT), to which their simpler nature also lends itself. The experiments amount to user studies questioning participants, with two different levels of expertise (the general population and those with some medical training), on their subjective and objective comprehension of and trust in explanations for the outputs of the regression tool. In both studies we found a clear preference in terms of subjective comprehension and trust for occlusion-1 over SHAP explanations in general, when comparing based on content. However, direct comparisons of explanations when controlling for format only revealed evidence for OT over SC explanations in most cases, suggesting that the dominance of occlusion-1 over SHAP explanations may be driven by a preference for text over charts as explanations. Finally, we found no evidence of a difference between the explanation types in terms of objective comprehension. Thus overall, the choice of the content and format of explanations needs careful attention, since in some contexts format, rather than content, may play the critical role in improving user experience., Comment: 18 pages

Metal halide perovskites (MHPs) have attracted significant attention owing to their simple manufacturing process and unique optoelectronic properties. Their reversible electrical or optical properties changes in response to oxidizing or reducing environments make them prospective materials for gas detection technologies. Despite advancements in perovskite-based sensor research, the mechanisms behind perovskite-gas interactions, vital for sensor performance, are still unexclusive. This work presents the first evaluation of the sensing performance and long-term stability of MHPs, considering factors such as halide composition variation and Mn doping levels. The research reveals a clear correlation between halide composition and sensing behavior, with Br-rich sensors displaying a p-type response to O3 gas, while Cl-based counterparts exhibit an n-type sensing behavior. Notably, Mn-doping significantly enhances the O3 sensing performance by facilitating the gas adsorption process, as supported by both atomistic simulations and experimental evidence. Long-term evaluation of the sensors provides valuable insights into evolving sensing behaviors, highlighting the impact of dynamic instabilities over time. Overall, this research offers insights into optimal halide combination and Mn-doping levels, representing a significant step forward in engineering room temperature perovskite-based gas sensors that are not only low-cost and high-performing but also durable, marking a new era in sensor technology., Comment: Main document 35 pages, 6 panel figures. Support information 28 pages, 43 figures

With the impending removal of third-party cookies from major browsers and the introduction of new privacy-preserving advertising APIs, the research community has a timely opportunity to assist industry in qualitatively improving the Web's privacy. This paper discusses our efforts, within a W3C community group, to enhance existing privacy-preserving advertising measurement APIs. We analyze designs from Google, Apple, Meta and Mozilla, and augment them with a more rigorous and efficient differential privacy (DP) budgeting component. Our approach, called Cookie Monster, enforces well-defined DP guarantees and enables advertisers to conduct more private measurement queries accurately. By framing the privacy guarantee in terms of an individual form of DP, we can make DP budgeting more efficient than in current systems that use a traditional DP definition. We incorporate Cookie Monster into Chrome and evaluate it on microbenchmarks and advertising datasets. Across workloads, Cookie Monster significantly outperforms baselines in enabling more advertising measurements under comparable DP protection., Comment: Published at SOSP '24. v5: typos and minor changes. v4: camera-ready version. v3: changed to non-anonymized name after acceptance notification, clarified text and reformatted graphs in {\S}8. v2: added pseudocode in {\S}3.3

This work reveals that the direct conjugation of metal halide perovskite nanocrystals on rGO sheets can provide high performance and stable electrodes for Zn-air supercapacitors. In particular, we provide a simple, rapid and room temperature laser-triggered method to anchor CsPbBr3 nanocrystals on rGO sheets without affecting the initial morphology and crystal structure of the components. The flexible and high surface area of the rGO enables the conjugation of individual metal halide perovskite nanocrystals with the emergence of new synergetic functionalities from the combination of the two counterparts. These synergetic functionalities lead to the 178- and 152-times improvement of the specific capacity of the conjugated-based electrodes compared to the single rGO and perovskite electrodes respectively., Comment: Main text 8 pages, 3 figures. Support information: 8 pages, 7 figures. Corresponding authors: Athanasia Kostopoulou - akosto@iesl.forth.gr, Dimitra Vernardou - dvernardou@hmu.gr, Emmanuel Stratakis - stratak@iesl.forth.gr

Differentially-private (DP) databases allow for privacy-preserving analytics over sensitive datasets or data streams. In these systems, user privacy is a limited resource that must be conserved with each query. We propose Turbo, a novel, state-of-the-art caching layer for linear query workloads over DP databases. Turbo builds upon private multiplicative weights (PMW), a DP mechanism that is powerful in theory but ineffective in practice, and transforms it into a highly-effective caching mechanism, PMW-Bypass, that uses prior query results obtained through an external DP mechanism to train a PMW to answer arbitrary future linear queries accurately and "for free" from a privacy perspective. Our experiments on public Covid19 and CitiBike datasets show that Turbo with PMW-Bypass conserves 1.7-15.9x more budget compared to vanilla PMW and simpler cache designs, a significant improvement. Moreover, Turbo provides support for range query workloads, such as timeseries or streams, where opportunities exist to further conserve privacy budget through DP parallel composition and warm-starting of PMW state. Our work provides a theoretical foundation and general system design for effective caching in DP databases., Comment: Extended version of a paper presented at the 29th ACM Symposium on Operating Systems Principles (SOSP '23)

Machine learning (ML) models can leak information about users, and differential privacy (DP) provides a rigorous way to bound that leakage under a given budget. This DP budget can be regarded as a new type of compute resource in workloads of multiple ML models training on user data. Once it is used, the DP budget is forever consumed. Therefore, it is crucial to allocate it most efficiently to train as many models as possible. This paper presents the scheduler for privacy that optimizes for efficiency. We formulate privacy scheduling as a new type of multidimensional knapsack problem, called privacy knapsack, which maximizes DP budget efficiency. We show that privacy knapsack is NP-hard, hence practical algorithms are necessarily approximate. We develop an approximation algorithm for privacy knapsack, DPack, and evaluate it on microbenchmarks and on a new, synthetic private-ML workload we developed from the Alibaba ML cluster trace. We show that DPack: (1) often approaches the efficiency-optimal schedule, (2) consistently schedules more tasks compared to a state-of-the-art privacy scheduling algorithm that focused on fairness (1.3-1.7x in Alibaba, 1.0-2.6x in microbenchmarks), but (3) sacrifices some level of fairness for efficiency. Therefore, using DPack, DP ML operators should be able to train more models on the same amount of user data while offering the same privacy guarantee to their users., Comment: Published at EuroSys '25. v2: camera-ready version

Background: Geospatial linked data brings into the scope of the Semantic Web and its technologies, a wealth of datasets that combine semantically-rich descriptions of resources with their geo-location. There are, however, various Semantic Web technologies where technical work is needed in order to achieve the full integration of geospatial data, and federated query processing is one of these technologies. Methods: In this paper, we explore the idea of annotating data sources with a bounding polygon that summarizes the spatial extent of the resources in each data source, and of using such a summary as an (additional) source selection criterion in order to reduce the set of sources that will be tested as potentially holding relevant data. We present our source selection method, and we discuss its correctness and implementation. Results: We evaluate the proposed source selection using three different types of summaries with different degrees of accuracy, against not using geospatial summaries. We use datasets and queries from a practical use case that combines crop-type data with water availability data for food security. The experimental results suggest that more complex summaries lead to slower source selection times, but also to more precise exclusion of unneeded sources. Moreover, we observe the source selection runtime is (partially or fully) recovered by shorter planning and execution runtimes. As a result, the federated sources are not burdened by pointless querying from the federation engine. Conclusions: The evaluation draws on data and queries from the agroenvironmental domain and shows that our source selection method substantially improves the effectiveness of federated GeoSPARQL query processing., Comment: Published in Open Research Europe

Sparse tensors appear frequently in distributed deep learning, either as a direct artifact of the deep neural network's gradients, or as a result of an explicit sparsification process. Existing communication primitives are agnostic to the peculiarities of deep learning; consequently, they impose unnecessary communication overhead. This paper introduces DeepReduce, a versatile framework for the compressed communication of sparse tensors, tailored for distributed deep learning. DeepReduce decomposes sparse tensors in two sets, values and indices, and allows both independent and combined compression of these sets. We support a variety of common compressors, such as Deflate for values, or run-length encoding for indices. We also propose two novel compression schemes that achieve superior results: curve fitting-based for values and bloom filter-based for indices. DeepReduce is orthogonal to existing gradient sparsifiers and can be applied in conjunction with them, transparently to the end-user, to significantly lower the communication overhead. As proof of concept, we implement our approach on Tensorflow and PyTorch. Our experiments with large real models demonstrate that DeepReduce transmits fewer data and imposes lower computational overhead than existing methods, without affecting the training accuracy.

Highly stable ozone and hydrogen sensing elements were fabricated based on well-crystalline rounded cube-shaped CsPbBr$_3$ submicron crystals, synthesized by a facile solution process performed under ambient conditions. It is shown that such elements demonstrate enhanced room temperature gas sensing ability compared to the previously reported metal halide and oxide-based ones. Electrical measurements performed on these sensing components revealed high response to ultra-low ozone and hydrogen concentrations, namely 4 ppb and 5 ppm respectively, as well as an impressive repeatability of the sensing behavior even after a few months of storage in ambient conditions. Both ozone and hydrogen detecting sensors were self-powered, i.e. they do not require the use of UV or heating external stimuli, and exhibited fast detection and short restoration times. These attractive properties along with the simple synthesis conditions could provide an easy, efficient and low-cost potential technology for the realization of future gas sensing devices., Comment: 16 pages, 5 figures, Support Information 10 pages, 7 figures, 2 tables

A facile, low-temperature precipitation-based method is utilized to demonstrate the synthesis of ultra-thin and highly-uniform cesium lead bromide perovskite nanowires (NWs). The reactions facilitate the NWs crystalline nature over micron-size lengths, while they impart tailored nanowire widths that range from the quantum confinement regime (~ 7 nm) and down to 2.6 nm. This colloidal synthesis approach is the first of its kind that is carried out on the work-bench, without demanding chemical synthesis equipment. Importantly, the NWs photoluminescence is shown to become improved over time, with no tedious post-synthesis surface treatment requirement., Comment: 9 pages, 3 Figures (main text) 10 pages, 10 Figures (supporting information)

In contrast to bulk materials, nanoscale crystal growth is critically influenced by size- and shape-dependent properties. However, it is challenging to decipher how stoichiometry, in the realm of mixed-valence elements, can act to control physical properties, especially when complex bonding is implicated by short and long-range ordering of structural defects. Here, solution-grown iron-oxide nanocrystals (NCs) of the pilot wustite system are found to convert into iron-deficient rock-salt and ferro-spinel sub-domains, but attain a surprising tetragonally distorted local structure. Cationic vacancies within chemically uniform NCs are portrayed as the parameter to tweak the underlying properties. These lattice imperfections are shown to produce local exchange-anisotropy fields that reinforce the nanoparticles magnetization and overcome the influence of finite-size effects. The concept of atomic-scale defect control in subcritical size NCs, aspires to become a pathway to tailor-made properties with improved performance for hyperthermia heating over defect-free NCs., Comment: 20 pages, 9 figures, Supplemental Material