We believe that transparency in real estate information results in more effective deal making. CompStak® is the market leader in commercial deal information (comps). Brokers exchange the comps they have at hand, for the specific comps they need.

CompStak

Are you wondering how you can elevate your business and keep up with all the advancements in the housing market? You’re not alone, industry leaders are integrating property technology to evolve in the 21st century. Property technology (Proptech) has the ability to elevate the real estate industry in all avenues, from purchase, maintenance, management, and even investment. It is now an industry that can assist with everything from property listing and rental services to mortgage applications, virtual and augmented reality modeling, Internet of Things (IoT) home devices, and property management software. $18 billion The Proptech industry is booming, and for good reason. In a survey of more than 100 real estate industry leaders, said that adopting new technologies was a strategic priority. two-thirds Reasons for this include effectively utilizing the environment, social, and governance (ESG) data, reducing costs, and the ability to access new revenue sources. Most industry leaders see property technology as an important tool, and the housing market is no exception. But what exactly is it? And how can these technologies change the market as we know it today? What is Proptech? Proptech takes information on platform economics and applies it to real estate markets. The goal of these technological innovations is to speed up transactions for those working in the housing market while making these transactions more secure and reducing the amount of paperwork required. The earliest form of Proptech is eCommerce; before the invention of the internet, it would be unheard of to rent or sell your property to complete strangers hundreds of miles away, but after decades of refining online marketplaces, the housing market is booming. Innovations in property technology have evolved into interlinked systems; they all have different functions but work together with like artificial intelligence (AI), virtual reality (VR), contract automation, real estate listings, data, the Internet of Things, construction technologies, and even chatbots! technologies The Rise of Proptech After the Pandemic The pandemic disrupted life and ushered in a new wave of online real estate tours that limit close contact, with VR home viewing and site visits skyrocketing in popularity and becoming a convenient solution for house-hunting during a modern-day plague. found that in 2020, walkthrough views where prospective buyers viewed the property online through a 3D scan increased a whopping 560%! Redfin.com Remote property appraisals have also become more normalized in the last few years, saving time, money, and cutting contact with people and their germs. There is also potential for VR to lead to an increase in sales of prefab buildings. It may be hard for a potential buyer to imagine the potential of a site, but with the use of VR, they would be able to see and experience of layouts, floor plans, and finishes. 3D representations Many companies Compass, Zillow, and Sotheby’s International Realty have integrated this technology into their home viewing models. including Where Do Start-ups Fit In? Start-ups are currently flooding the real estate market with innovative ideas and creative solutions. With this new wave, the industry is seeing an influx of changes when it comes to management experience and purchasing properties. Some of these include operating buildings in a more sustainable manner, eliminating the middleman from property purchasing, and giving tenants data and insights on how they use their space and shortcuts they can take to more efficiently save power and energy. changes Proptech is implemented by all kinds of start-ups in the real estate industry, property owner analytics, building efficiency data, and researching market analytics. including One such company is CompStak, which just received in funding to crowdsource and compile detailed data on office, industrial, and retail lease and sales terms. Currently, they have 1.7 million properties in their database. And they are constantly expanding. $50 million is a Proptech app that allows building managers to quickly and efficiently notify tenants of issues. The app also collects and analyzes occupancy data that give property owners a cohesive understanding of how the space is used. This can help optimize the space, as well as inform renovation decisions. Spaceti Another example of a Proptech-based start-up is . Perfect for Airbnb hosts, as well as anyone else who rents out their home or office building, the igloohome technology works to add security and ease to the experience. igloohome Users can their Airbnb calendar with their igloohome lock, which will automatically generate custom PIN codes only valid for the duration of the stay and send them to the guest(s). sync The real estate industry is one of the largest in the world but is lagging behind when it comes to technological implementation. The goal is to add to the existing formula and to make cumbersome transactions easier, one way or another. Implementing new strategies, models, and technologies can be intimidating, but it's pertinent for the housing industry to evolve and adapt. , Co-founder of Chris Pchelintsev Architecturalist

Proptech: The Evolution of the Housing Market

Benchmarking the Computational Performance of Poseidon2 in Plonky3

Authors:
(1) Renan F. F. da Silva, Institute of Computing, University of Campinas;
(2) Yulle G. F. Borges, Institute of Computing, University of Campinas;
(3) Rafael C. S. Schouery, Institute of Computing, University of Campinas. 8 Computational Results Next, we present the instances used in our experiments [1] . Sequentially, we show some results comparing our algorithms in two-stage. We compare our initial heuristics using a performance profile and statistic tests in the first stage. In the second stage, we make a statistical comparison between the VNS and the MH. 8.1 Instances The instances used in experiments have four classes. The BPP instances AI and FalkenauerT are available in the BPP Lib created by Delorme et al [15], and their optimal solutions are known. In the BPP, AI instances are very hard instances to find an optimal solution, but they are easy to find a solution with an extra bin. The instances FalkenauerT  and Random25to50 are similar, but we create the Random25to50 because FalkenauerT does not  have instances with a large number of items or a large bin capacity. 8.2 Comparison between Initial Heuristics Next, we compare our initial heuristics using a performance profile and statistic tests. 8.2.1 Performance Profile In Figure 1, we have the performance profile for the combinations of AI instances with the coloring Q2 and Half Q2. As we can see, the heuristic Two-by-Two is the best, being up to two times better than the BFD in some instances in coloring Half Q2. The graph for the other coloring is similar to Q2, but the difference decreases as we increase the number of colors. This behavior also happens in other instances, so we omit the graphs for these coloring. Figure 2 presents the results for Random25to50 instances with coloring Q2 and Half Q2. Again, the heuristic Two-by-Two is the best one. This time, it has a reasonable superiority to Hard-BFD, which use at least 10% more bins. See that the two MBS’ versions are curiously better than the Hard-BFD in 90% of the instances with Q2. The results for the FalkenauerT instances are similar to Random25to50. Next, in Figure 3, we have the performance profile for Random10to80 with coloring Q2 and Q5. Observe that with the coloring Q2, we do not have a clear dominance between the heuristics Twoby-Two and Hard-BFD. Finally, note that the good-ordering often improves the solution quality, but exceptions exist. For example, the Good Ordering heuristic is worse than the BFD heuristic in AI instances with Q2. 8.2.2 Statistic Tests The comparison we made is based on non-parametric analysis, as was made by Demˇsar [16]. As we can compare multiple strategies on a set of different instances, we use the adaptation of the Friedman test (Friedman [20]) made by Iman and Davenport [24], rejecting the null hypothesis when the value of the test is greater than the critical one according to the F-distribution. In such a case, we proceed with a post hoc analysis using the Nemenyi test (Nemenyi [29]) to identify the group of statistically equivalent strategies. This situation is identified when the average ranks of two samples differ at most by a given critical difference (CD). Our test has k = 6 algorithms and N = 1200 instances. For these parameters, the critical value of the F-distribution is 2.22. Next, we can compute the test value FF to the Iman-Davenport, where our data has FF = 484.02. As FF is greater than the critical value, the null hypothesis H0 was rejected, and we accepted the alternative hypothesis H1. Therefore, we can apply the Nemenyi test to determine which heuristics have statistical differences. Figure 4 presents a graph illustrating the Nemenyi test. For our data, we have the critical difference CD = 0.22. The heuristic average ranks are classified from left to right, where the better heuristic is the most on the left. If two heuristics have average ranks with a difference less than the CD, then they are connected by a horizontal line, which means that we did not find a statistical difference between them. As we can see in the figure, the Two-by-Two is the best heuristic. Note that the “good-ordering” is beneficial in the general case since MBS’-GO and GO heuristics perform better than their original versions MBS’ and BFD. Finally, we do not find a statistical difference between MBS’ and GO heuristics. 8.3 Comparison between VNS and MH In the Tables 1 and 2, the columns “Initial Avg. Diff.” e “Final Avg. Diff.” indicate the average differences between the lower bound L1 and the initial and final solution values, respectively. The column “Final Max. Diff.” indicate the maximum difference of a final solution about lower bound L1. Moreover, the column “Time” gives the average time in seconds. Finally, we highlight in bold the best “Final Avg. Diff.” of each line. As we can see, our approaches obtained good results since they optimally solve all Random25to50 instances with 102 and 501 items and find solutions with at most one extra bin than the optimal solution with 2001 items. Moreover, we can emphasize this class with 10002 items and coloring Half Q2, where the average initial difference is 82.1 bins, and our approaches decrease this difference to at most 1 bin. Besides, our approaches also solve all FalkenauerT (FK) instances. Note that there is no clear relation between difficulty and the number of items since instances with 120 items and Half Q2 are the most costly in execution time. Moreover, see that the coloring Half Q2 has the greater initial average difference in all instance classes. However, it is not necessarily reflected in more hard instances since in some cases, the final maximum difference is slightly smaller in this coloring. Finally, our MH is superior to VNS since the MH optimally solves 124, 70, and 9 of the 150 AI instances with 202, 403, and 601 items, respectively, while the VNS does not solve any of them. Moreover, in some cases, the average final difference in Random10to80 instances is several times smaller in MH than in VNS. Note that the average difference in the final solutions is less than 3 in the MH. This difference is significantly greater in Random10to80 instances, where L1 is probably not a tight lower bound. [1] All instances and source codes are available at https://gitlab.com/renanfernandofranco/ fast-neighborhood-search-heuristics-for-the-colorful-bin-packing-problem. This paper is available on arxiv under CC 4.0 license. Authors: (1) Renan F. F. da Silva, Institute of Computing, University of Campinas; (2) Yulle G. F. Borges, Institute of Computing, University of Campinas; (3) Rafael C. S. Schouery, Institute of Computing, University of Campinas. Authors: (1) Renan F. F. da Silva, Institute of Computing, University of Campinas; (2) Yulle G. F. Borges, Institute of Computing, University of Campinas; (3) Rafael C. S. Schouery, Institute of Computing, University of Campinas. 8 Computational Results Next, we present the instances used in our experiments [1] . Sequentially, we show some results comparing our algorithms in two-stage. We compare our initial heuristics using a performance profile and statistic tests in the first stage. In the second stage, we make a statistical comparison between the VNS and the MH. 8.1 Instances The instances used in experiments have four classes. The BPP instances AI and FalkenauerT are available in the BPP Lib created by Delorme et al [15], and their optimal solutions are known. In the BPP, AI instances are very hard instances to find an optimal solution, but they are easy to find a solution with an extra bin. The instances FalkenauerT  and Random25to50 are similar, but we create the Random25to50 because FalkenauerT does not  have instances with a large number of items or a large bin capacity. 8.2 Comparison between Initial Heuristics Next, we compare our initial heuristics using a performance profile and statistic tests. 8.2.1 Performance Profile In Figure 1, we have the performance profile for the combinations of AI instances with the coloring Q2 and Half Q2. As we can see, the heuristic Two-by-Two is the best, being up to two times better than the BFD in some instances in coloring Half Q2. The graph for the other coloring is similar to Q2, but the difference decreases as we increase the number of colors. This behavior also happens in other instances, so we omit the graphs for these coloring. Figure 2 presents the results for Random25to50 instances with coloring Q2 and Half Q2. Again, the heuristic Two-by-Two is the best one. This time, it has a reasonable superiority to Hard-BFD, which use at least 10% more bins. See that the two MBS’ versions are curiously better than the Hard-BFD in 90% of the instances with Q2. The results for the FalkenauerT instances are similar to Random25to50. Next, in Figure 3, we have the performance profile for Random10to80 with coloring Q2 and Q5. Observe that with the coloring Q2, we do not have a clear dominance between the heuristics Twoby-Two and Hard-BFD. Finally, note that the good-ordering often improves the solution quality, but exceptions exist. For example, the Good Ordering heuristic is worse than the BFD heuristic in AI instances with Q2. 8.2.2 Statistic Tests The comparison we made is based on non-parametric analysis, as was made by Demˇsar [16]. As we can compare multiple strategies on a set of different instances, we use the adaptation of the Friedman test (Friedman [20]) made by Iman and Davenport [24], rejecting the null hypothesis when the value of the test is greater than the critical one according to the F-distribution. In such a case, we proceed with a post hoc analysis using the Nemenyi test (Nemenyi [29]) to identify the group of statistically equivalent strategies. This situation is identified when the average ranks of two samples differ at most by a given critical difference (CD). Our test has k = 6 algorithms and N = 1200 instances. For these parameters, the critical value of the F-distribution is 2.22. Next, we can compute the test value FF to the Iman-Davenport, where our data has FF = 484.02. As FF is greater than the critical value, the null hypothesis H0 was rejected, and we accepted the alternative hypothesis H1. Therefore, we can apply the Nemenyi test to determine which heuristics have statistical differences. Figure 4 presents a graph illustrating the Nemenyi test. For our data, we have the critical difference CD = 0.22. The heuristic average ranks are classified from left to right, where the better heuristic is the most on the left. If two heuristics have average ranks with a difference less than the CD, then they are connected by a horizontal line, which means that we did not find a statistical difference between them. As we can see in the figure, the Two-by-Two is the best heuristic. Note that the “good-ordering” is beneficial in the general case since MBS’-GO and GO heuristics perform better than their original versions MBS’ and BFD. Finally, we do not find a statistical difference between MBS’ and GO heuristics. 8.3 Comparison between VNS and MH In the Tables 1 and 2, the columns “Initial Avg. Diff.” e “Final Avg. Diff.” indicate the average differences between the lower bound L1 and the initial and final solution values, respectively. The column “Final Max. Diff.” indicate the maximum difference of a final solution about lower bound L1. Moreover, the column “Time” gives the average time in seconds. Finally, we highlight in bold the best “Final Avg. Diff.” of each line. As we can see, our approaches obtained good results since they optimally solve all Random25to50 instances with 102 and 501 items and find solutions with at most one extra bin than the optimal solution with 2001 items. Moreover, we can emphasize this class with 10002 items and coloring Half Q2, where the average initial difference is 82.1 bins, and our approaches decrease this difference to at most 1 bin. Besides, our approaches also solve all FalkenauerT (FK) instances. Note that there is no clear relation between difficulty and the number of items since instances with 120 items and Half Q2 are the most costly in execution time. Moreover, see that the coloring Half Q2 has the greater initial average difference in all instance classes. However, it is not necessarily reflected in more hard instances since in some cases, the final maximum difference is slightly smaller in this coloring. Finally, our MH is superior to VNS since the MH optimally solves 124, 70, and 9 of the 150 AI instances with 202, 403, and 601 items, respectively, while the VNS does not solve any of them. Moreover, in some cases, the average final difference in Random10to80 instances is several times smaller in MH than in VNS. Note that the average difference in the final solutions is less than 3 in the MH. This difference is significantly greater in Random10to80 instances, where L1 is probably not a tight lower bound. [1] All instances and source codes are available at https://gitlab.com/renanfernandofranco/ fast-neighborhood-search-heuristics-for-the-colorful-bin-packing-problem. This paper is available on arxiv under CC 4.0 license. This paper is available on arxiv under CC 4.0 license. available on arxiv

Analyzing Computational Results: Insights into Bin Packing Heuristic Performance

How Useful are Computational Models to Traders in Their Decision Making?

Authors: (1) JunJie Wee, Department of Mathematics, Michigan State University; (2) Jiahui Chen, Department of Mathematical Sciences, University of Arkansas; (3) Kelin Xia, Division of Mathematical Sciences, School of Physical and Mathematical Sciences Nanyang Technological University & xiakelin@ntu.edu.sg; (4)Guo-Wei Wei, 1Department of Mathematics, Michigan State University, Department of Biochemistry and Molecular Biology, Michigan State University, Department of Electrical and Computer Engineering, Michigan State University & xiakelin@ntu.edu.sg. Table of Links Abstract & Introduction Results Discussion Conclusion Materials and Methods Software and resources, Code and Data Availability Supporting Information, Acknowledgments & References 5 Materials and Methods In this section, we endeavor to elucidate key mathematical and computational foundations that are instrumental for the work presented in this study. Specifically, we delve into spectral graph theory, simplicial complex, and persistent Laplacian methods, highlighting their significance in capturing topological and spectral properties essential for the characterization of proteins. Additionally, we discuss machine learning and deep learning paradigms, focusing on their role in processing, analyzing, and interpreting these complex features, especially within the confines of test datasets and validation settings. 5.1 Persistent Laplacian characterization of proteins A simplicial complex is made up of a set of simplices and generalises beyond graph networks at higher dimensions [44, 45, 46, 47]. Every simplex is a finite set of vertices which can be interpreted as the atoms in a protein structure. Essentially, simplices can be a point (0-simplex), an edge (1-simplex), a triangle (2-simplex), a tetrahedron (3-simplex), or in higher dimensions, a p-simplex. In other words, a k-simplex σ k = {v0, v1, · · · , vk} is the convex hull formed by k + 1 affinely independent points v0, v1, · · · , vk as follows, Simplicial complex A geometric simplicial complex K is a finite set of geometric simplexes that satisfy two essential conditions. First, any face of a simplex from K is also in K. Second, the intersection of any two simplexes in K is either empty or shares faces. Commonly used methods to construct simplicial complexes are Cech complex, Vietoris-Rips complex, Alpha complex, Clique ˇ complex, Cubic complex, and Morse complex [44, 45, 46, 47]. In the case k = 0, then L0 = B1B⊤ 1 since ∂0 is a zero map Persistent Laplacian (PL) were first introduced by integrating graph Laplacian and multiscale filtration [25]. Analyzing the spectra of k-combinatorial Laplacian matrix allows both topological and geometric information (i.e. connectivity and robustness of simple graphs) to be obtained. However, this method is genuinely free of metrics or coordinates, which induced too little topological and geometric information that can be used to describe a single configuration. Persistent Laplacian Therefore, PL was extended to simplicial complexes. This allows a sequence of simplicial complexes from a filtration process to generate persistent Laplacian which is largely inspired by persistent homology and in earlier works in multiscale graphs. For the rest of this section, we introduce mainly on the construction of PL. First, a k-combinatorial Laplacian matrix is symmetric and positive semi-definite. Therefore, its eigenvalues are all real and non-negative. The multiplicity of zero spectra (also called harmonic spectra) reveals the topological information, and the geometric information will be preserved in the non-harmonic spectra. This nested sequence of simplicial complexes induces a family of chain complexes In order to illustrate the difference between PL and PH, Figure 5 describes a point cloud, basic simplices, a filtration process and the comparison between persistent Laplacian and persistent homology barcodes of 13 points. The filtration process in Figure 5(c) shows the different stages of a Rips filtration process for the 13 points. Figure 5(d) shows the persistent homology barcodes (in blue) and persistent non-harmonic spectra (in red). It can be seen that the nonharmonic spectra provides the additional homotopic shape evolution that is missing in persistent homology in the later part of the filtration process. 5.2 Persistent Laplacian descriptors where DE(·, ·) is the Euclidean distance between the two atoms and Loc(·) refers to the atom’s location which is either in the mutation site or in the rest of the protein. Here, we construct two types of simplicial complexes in our PL computation, such as Vietoris-Rips complex (VC) and Alpha complex (AC). Both complexes are used to characterize the first order interactions and higher order patterns respectively. To capture and characterize different types of atomatom interactions, we generate the PL based on different atom subsets by selecting one type of atom in the mutation site and one other atom type in the rest of the protein. Different types of atom-atom interactions characterize the different interactions in proteins. For example, interactions generated from carbon atoms are associated with hydrophobic interactions. Similarly, interactions between nitrogen and oxygen atoms correlate to hydrophilic interactions and/or hydrogen bonds. Both types of interactions are illustrated in Figure 6. Interactive PLs have the capability to unveil additional details about bonding interactions and offer a fresh and distinct representation of molecular interactions in proteins. For zero dimensions, we consider both the harmonic spectra and non-harmonic spectra information for each persistent Laplacian. Filtration using Rips complex with DI distance is used. The 0-dimensional PL features are generated from 0A to 6 ˚ A with 0.5 ˚ A gridsize. For the ˚ non-harmonic spectra information, we count the number of non-harmonic spectra and calculate seven statistical values of non-harmonic spectra such as sum, minimum, maximum, mean, standard deviation, variance and the sum of eigenvalues squared. This generates eight statistical values for each of the nine atomic pairs. Therefore, the dimension of 0-dimensional PL features for a protein is 72. In total, the 0-dimensional PL-based feature size after concatenating features at different dimensions for wild type and mutant is 1872. For one or two dimensions, we perform the filtration using Alpha complex with the DE distance. The limited number of atoms in the local protein structure can create only a few high-dimensional simplexes, resulting in minimal alterations in shape. As a result, it suffice to consider features from only harmonic spectra of persistent Laplacians by coding the topological invariants for the high-dimensional interactions. Using GUDHI[51], the persistence of the harmonic spectra can be represented by persistent barcodes. The topological feature vectors are generated by computing the statistics of bar lengths, births and deaths. Bars shorter than 0.1A are excluded as they do not exhibit any clear physical meaning. The remaining bars are ˚ then used for computing the statistics: (1) sum, maximum and mean for lengths of bars; (2) minimum and maximum for the birth values of bars; (3) minimum and maximum for the death values of bars. Each set of point clouds leads to a seven-dimensional vector. These features are calculated on nine single atomic pairs and one heavy atom pair. The dimension of oneand two-dimensional PL feature vectors for a protein is 140. In total, the higher-dimensional PL-based feature size after concatenating features at different dimensions for wild type, mutant and their difference is 420. 5.3 Persistent Homology Persistent homology is part of the harmonic spectra of PL. The homology groups in PH illustrate the persistence of topological invariants, hence providing the harmonic spectral information in PL. The site- and element-specific PH features are generated in a similar way as compared to PL. Similar to PL, filtration construction is also employed to PH. For the zero dimension, the filtration parameter can be discretized into several equally spaced bins, namely [0, 0.5], (0.5, 1], · · · , (5.5, 6]A. The death value of the bars are summed in each bin resulting in 12 ˚ ×18 features. For each bin, we count the numbers of persistent bars, resulting in a nine-dimensional vector for each point cloud. Similarly, this is performed for each of the nine single atomic pairs. Hence, the dimension of PH features for a protein is 216. For one or two dimensions, the identical featurization from the statistics of persistent bars in PH is used. The PH embedding combines features at different dimensions as described above and concatenated for wild type, mutant and their difference, resulting in a 648-dimensional vector 5.4 Transformer Features Recently, we have seen significant advancements in modelling protein properties using largescale protein transformer models trained on hundreds of millions of sequences. These models, like ESM [33] (evolutionary scale modeling) and ProtTrans[34, 35], have demonstrated impressive performance. Moreover, hybrid fine-tuning approaches that leverage both local and global evolutionary data have proven to enhance these models even further. For instance, eUniRep is an improved LSTM-based UniRep model achieved through fine-tuning with knowledge extracted from local multiple sequence alignments (MSAs). Additionally, the ESM model can be fine-tuned using either downstream task data or local MSAs. In our research, we employed the ESM-1b transformer, a model that falls under the transformer architecture. This particular variant was trained on a dataset of 250 million sequences using a masked filling procedure and boasts an architecture comprising 34 layers with a whopping 650 million parameters. The ESM transformer’s primary role in our work was to generate sequence embeddings. At each layer of the ESM model, it encoded a sequence of length L into a matrix sized at 1,280×L, excluding the start and terminal tokens. For our study, we utilized the sequence representation derived from the final (34th) layer and computed the average along the sequence length axis, resulting in a 1,280-component vector. 5.5 Performance Metrics PPV and NPV assesses the true positive and true negative proportion of the predicted results for each solubility class. PPV and NPV are computed based on TP, TN, FP and FN which represents the true positive, true negative, false positive and false negative values for each solubility class. For each solubility class, PPV and NPV can be computed by: This paper is under CC 4.0 license. available on arxiv

Interrogating TopLabGBT's Computational Framework

This paper is under CC 4.0 license. available on arxiv Authors: (1) Tariq Bontekoe, University of Groningen; (2) Dimka Karastoyanova, University of Groningen; (3) Fatih Turkmen, University of Groningen. Table of Links Abstract & Introduction Preliminaries Zero-knowledge proofs and verifiable computing Privacy-preserving computations Requirements: an application’s perspective Verifiable, privacy-preserving computing Open challenges and future directions Conclusion & References 4. Privacy-preserving computations There are different PETs that can be used to construct solutions for privacy-preserving computations. Two of them are specifically suitable for our setting: MPC and HE. Both are well-established PETs, providing flexibility and offering strong privacy guarantees, which make them excellent building blocks for verifiable privacy-preserving computations. Below we discuss both building blocks, and provide an intuitive answer to . RQ1 4.1. Secure multi-party computation MPC is a collection of techniques that allows for performing joint computations over distributed data, while keeping all input data private and revealing nothing but the computation output. Generally, there are n parties that participate in an MPC computation, and any participating party contributes to the collaboration with their respective private inputs. All n parties execute an interactive protocol to compute the joint function output from their private inputs. An MPC protocol must therefore be resistant to the fact that some participating parties may be corrupted by an adversary. As discussed in Section 2.2, there exist different security models. However, any secure MPC protocol should at least satisfy the following (informal) requirements [14], given the constraints of the used adversarial model: • Input privacy: No party should learn anything other than the function output it is assigned. Specifically, no party should learn anything about other parties’ inputs, other than what can be derived from its assigned output. • Correctness: Every party that receives a (partial) output, obtains a value that is correct with respect to the protocol specification and input data. • Independence of inputs: Corrupted parties must pick their inputs independently of the honest parties’ private inputs. There are several ways to construct an MPC protocol or framework. Below, we present the two most common constructions: secret sharing and garbled circuits. In a secret sharing approach, private data is split over multiple parties. Information is split in such a way that only predefined sets of parties can reconstruct the full information, while other (sub)sets cannot even deduce partial information. Most secret sharing schemes divide a secret s among n parties in such a way that only subsets with t or more parties can reconstruct s. Such a scheme is called a t-out-of-n-threshold scheme. Secret Sharing. Finally, we note that shares from an LSSS can be multiplied efficiently using so called Beaver’s multiplication triplets [30]. These triplets are input/function-independent and can thus be generated in a so called offline preprocessing phase. Using this approach, the remaining computations can be performed in a very efficient online phase. Many popular, modern secret sharing schemes make use of linear shares in combination with Beaver triplets or oblivious transfer. Example schemes using secret sharing are: Shamir’s secret sharing (SSS) [31] or, in the online phase, SPDZ [32] and MASCOT [33]. For an up-to-date overview on existing schemes with implementations we refer to the MP-SPDZ library [34]. Yao’s garbled circuits (GCs) [35] are a way to enable two distrusting parties to securely evaluate a function on their private input data. It requires that the underlying function is expressed as a boolean circuit consisting of 1-out-2-in gates. Garbled Circuits. One party, the garbler, generates a garbled version of the entire circuit, by garbling the individual gates. In the original implementation, a gate is garbled by assigning a unique, random label to the possible values (true, false) of each input wire, and doubly encrypting the four possible output values, under the corresponding input labels. The garbler randomly permutes the encrypted outputs and shares these with the evaluator accompanied by the random labels corresponding to the private inputs of the garbler. The evaluator then participates in an oblivious transfer (OT) protocol [36] to obtain the labels corresponding to their private inputs, without revealing those. Having received both input labels, the evaluator can correctly decrypt exactly one of the garbled outputs, thereby obtaining the true output bit(s). When a gate output is used as input to another gate, the true output will not be a bit, but rather a new random label that is used in the decryption of this other gate. An alternative construction to Yao’s GCs is the BMR framework [37]. Implementations of both and other frameworks can be found in, e.g., MP-SPDZ [34], ABY [38], or ABY 3 [39]. Most MPC frameworks are applied in a fully decentralized setting, where all parties communicate with one another directly. Many MPC schemes guarantee privacy for a given threat model, under the assumption of a certain number of honest parties, ranging from a strict majority to only one. Therefore, one might say that correctness is a largely solved problem. RQ1a: Why should we verify MPC? However, none of the above takes into account the use case in which parties that did not participate in the computation want to verify the results. In this situation, the results could be incorrect when all computing parties are colluding. The schemes are thus not publicly verifiable or publicly auditable [40], making the results untrustworthy for external parties. Another issue regarding (public) verifiability, that is often overlooked, is verification of data authenticity. Correctness of results is often only guaranteed given the assumption that input data is provided honestly. This especially plays a role when input data of parties should be the same over multiple computations, e.g., for reproducibility, or when input data should be authenticated, e.g., signed by an expert authority. In practice, both external and computation parties that want verifiably correct results, will require verification of computation and data authenticity. 4.2. Homomorphic encryption HE is a term used to denote a collection of (asymmetric) encryption schemes that allow users to perform operations on encrypted data without decrypting it first. More formally, and following [41], an HE scheme is described by four (probabilistic) polynomial-time algorithms (KeyGen, Enc, Dec, Eval): • KeyGen(1κ ) → (sk, pk, evk) is the key generation algorithm. It takes a unary representation of the security parameter κ as input and outputs a secret decryption key sk, public encryption key pk, and public evaluation key evk. While there are different ways to represent the function f in HE schemes, most constructions prefer a translation to an arithmetic circuit Cf over M. Notably, not all homomorphic encryption schemes can be used to evaluate arbitrary arithmetic circuits. Therefore, we distinguish, following [42], several types of homomorphic schemes in increasing order of evaluation capabilities. PHE is a type of homomorphic encryption that is only additively or only multiplicatively homomorphic, implying that it can only evaluate arithmetic circuits consisting solely of addition, respectively multiplication gates. In general, partially homomorphic schemes have the lowest computational cost and stable ciphertext size. Partially homomorphic encryption. SWHE constructions can evaluate arithmetic circuits consisting of both addition and multiplication gates. However, the circuits that can be evaluated may be limited in the number of operations (of either type), for example a scheme that can only evaluate circuits with a maximum multiplication depth of 2. Moreover, for SWHE there is no requirement that prevents exponential blowup of the ciphertext size.. Somewhat homomorphic encryption. LHE resembles SWHE in that it can evaluate circuits that are limited in depth d. However, since d is provided as input to KeyGen, the scheme parameters may be a function of d. Next to this, the length of the ciphertext output of the Eval algorithm may not depend on the depth d, meaning that ciphertexts do not blow up exponentially. Notice that the key sizes may still depend on the depth d and can get impractically large. Leveled homomorphic encryption. FHE schemes can evaluate arbitrary arithmetic circuits, without the schemes parameters depending upon any circuit property. This generality often comes at the cost of expensive bootstrapping algorithms after a certain number of multiplications [43]. Fully homomorphic encryption. HE is often used in a setting where one (cloud) server or central entity performs the majority of computations, for example in an outsourcing scenario. A frequently observed scenario is where all data owners use a distributed, or threshold, key generation protocol to obtain a public-private key pair where the private key is shared over all data owners [44], [45]. To perform a computation, all data owners encrypt their data under the same public key and let the central server perform the actual computation on the received ciphertexts. Due to the high computational cost of operations on homomorphically encrypted data, it is often preferred to have a single, central computing server with sufficient computational resources. RQ1b: Why should we verify HE? In most cases all parties are assumed to behave semihonestly. This assumption, however, is unrealistic in most real-world scenarios and raises significant integrity and verifiability issues. Integrity issues arise from the possibility of key-recovery attacks [46], [47], [48], whereas verifiability issues stem from the fact that the server can deviate from the requested computation. Another issue regarding verifiability, is whether external parties can verify the results of a computation that uses HE. These guarantees may be warranted in cases where external auditing is required or when results are of interest to a larger community. Therefore, verifiable solutions should provide guarantees even in the case where all data owners and the central computation server(s) are colluding. Next to this, external parties may also require guarantees concerning the integrity and authenticity of input data, which is in general not considered in existing work.

Understanding Privacy-preserving Computations

This paper is under CC BY 4.0 DEED license. available on arxiv Authors: (1) Ehsan Toreini, University of Surrey, UK; (2) Maryam Mehrnezhad, Royal Holloway University of London; (3) Aad Van Moorsel, Birmingham University. Table of Links Abstract & Introduction Background and Related Work FaaS Architecture Implementation and Performance Analysis Conclusion Acknowledgment and References Acknowledgment The authors in this project have been funded by UK EPSRC grant “FinTrust: Trust Engineering for the Financial Industry” under grant number EP/R033595/1, and UK EPSRC grant “AGENCY: Assuring Citizen Agency in a World with Complex Online Harms” under grant EP/W032481/1 and PETRAS National Centre of Excellence for IoT Systems Cybersecurity, which has been funded by the UK EPSRC under grant number EP/S035362/1. References [1] Philip Adler, Casey Falk, Sorelle A Friedler, Tionney Nix, Gabriel Rybeck, Carlos Scheidegger, Brandon Smith, and Suresh Venkatasubramanian. Auditing black-box models for indirect influence. Knowledge and Information Systems, 54(1):95–122, 2018. [2] Elaine Angelino, Nicholas Larus-Stone, Daniel Alabi, Margo Seltzer, and Cynthia Rudin. Learning certifiably optimal rule lists for categorical data. arXiv preprint arXiv:1704.01701, 2017. [3] Muhammad Ajmal Azad, Samiran Bag, Simon Parkinson, and Feng Hao. Trustvote: Privacy-preserving node ranking in vehicular networks. IEEE Internet of Things Journal, 6(4):5878–5891, 2018. [4] Olivier Baudron, Pierre-Alain Fouque, David Pointcheval, Jacques Stern, and Guillaume Poupard. Practical multi-candidate election system. In Proceedings of the twentieth annual ACM symposium on Principles of distributed computing, pages 274–283, 2001. [5] Alexandra Chouldechova. Fair prediction with disparate impact: A study of bias in recidivism prediction instruments. Big data, 5(2):153–163, 2017. [6] Sam Corbett-Davies, Emma Pierson, Avi Feller, Sharad Goel, and Aziz Huq. Algorithmic decision making and the cost of fairness. In Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pages 797–806. ACM, 2017. [7] Ronald Cramer, Ivan Damg˚ard, and Berry Schoenmakers. Proofs of partial knowledge and simplified design of witness hiding protocols. In Annual International Cryptology Conference, pages 174–187. Springer, 1994. [8] Craig E. Carroll and Rowena Olegario. Pathways to corporate accountability: Corporate reputation and its alternatives. Journal of Business Ethics, 163(2):173–181, 2020. [9] Michael Feldman, Sorelle A Friedler, John Moeller, Carlos Scheidegger, and Suresh Venkatasubramanian. Certifying and removing disparate impact. In Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pages 259–268. ACM, 2015. [10] Amos Fiat and Adi Shamir. How to prove yourself: Practical solutions to identification and signature problems. In Conference on the theory and application of cryptographic techniques, pages 186–194. Springer, 1986. [11] Sorelle A Friedler, Carlos Scheidegger, and Suresh Venkatasubramanian. On the (im) possibility of fairness. arXiv preprint arXiv:1609.07236, 2016. [12] Riccardo Guidotti, Anna Monreale, Salvatore Ruggieri, Franco Turini, Fosca Giannotti, and Dino Pedreschi. A survey of methods for explaining black box models. ACM computing surveys (CSUR), 51(5):1–42, 2018. [13] Feng Hao, Matthew N Kreeger, Brian Randell, Dylan Clarke, Siamak F Shahandashti, and Peter Hyun-Jeen Lee. Every vote counts: Ensuring integrity in large-scale electronic voting. In 2014 Electronic Voting Technology Workshop/Workshop on Trustworthy Elections (EVT/WOTE 14), 2014. [14] Feng Hao, Peter YA Ryan, and Piotr Zieli´nski. Anonymous voting by two-round public discussion. IET Information Security, 4(2):62–67, 2010. [15] Moritz Hardt, Eric Price, Nati Srebro, and others. Equality of opportunity in supervised learning. In Advances in neural information processing systems, pages 3315–3323, 2016. [16] Hui Hu, Yijun Liu, Zhen Wang, and Chao Lan. A distributed fair machine learning framework with private demographic data protection. In 2019 IEEE International Conference on Data Mining (ICDM), pages 1102–1107. IEEE, 2019. [17] Matthew Jagielski, Michael Kearns, Jieming Mao, Alina Oprea, Aaron Roth, Saeed Sharifi-Malvajerdi, and Jonathan Ullman. Differentially private fair learning. In International Conference on Machine Learning, pages 3000–3008. PMLR, 2019. [18] Jonathan Katz and Yehuda Lindell. Introduction to modern cryptography. CRC press, 2014. [19] Niki Kilbertus, Adria Gascon, Matt Kusner, Michael Veale, Krishna P Gummadi, and Adrian Weller. Blind justice: Fairness with encrypted sensitive attributes. In 35th International Conference on Machine Learning, pages 2630–2639. PMLR, 2018. [20] Jeff Larson, Surya Mattu, Lauren Kirchner, and Julia Angwin. How we analyzed the COMPAS recidivism algorithm. ProPublica (5 2016), 9(1), 2016. [21] Jiacheng Liu, Fei Yu, and Lixin Song. A systematic investigation on the research publications that have used the medical expenditure panel survey (MEPS) data through a bibliometrics approach. Library Hi Tech, 2020. [22] Scott M Lundberg and Su-In Lee. A unified approach to interpreting model predictions. Advances in neural information processing systems, 30, 2017. [23] Arwa Mahdawi. It’s not just A-levels – algorithms have a nightmarish new power over our lives. The Guardian, 2020. [24] Arvind Narayanan. Translation tutorial: 21 fairness definitions and their politics. In Proc. Conf. Fairness Accountability Transp., New York, USA, 2018. [25] Cecilia Panigutti, Alan Perotti, Andr´e Panisson, Paolo Bajardi, and Dino Pedreschi. Fairlens: Auditing black-box clinical decision support systems. Information Processing & Management, 58(5):102657, 2021. [26] Cecilia Panigutti, Alan Perotti, and Dino Pedreschi. Doctor xai: an ontology-based approach to black-box sequential data classification explanations. In Proceedings of the 2020 conference on fairness, accountability, and transparency, pages 629–639, 2020. [27] Saerom Park, Seongmin Kim, and Yeon-sup Lim. Fairness audit of machine learning models with confidential computing. In Proceedings of the ACM Web Conference 2022, pages 3488–3499, 2022. [28] Reuters. Amazon ditched AI recruiting tool that favored men for technical jobs. The Guardian, 2018. [29] Shahar Segal, Yossi Adi, Benny Pinkas, Carsten Baum, Chaya Ganesh, and Joseph Keshet. Fairness in the eyes of the data: Certifying machinelearning models. In Proceedings of the 2021 AAAI/ACM Conference on AI, Ethics, and Society, pages 926–935, 2021. [30] Keng Siau and Weiyu Wang. Building trust in artificial intelligence, machine learning, and robotics. Cutter business technology journal, 31(2):47– 53, 2018. [31] Douglas Robert Stinson and Maura Paterson. Cryptography: theory and practice. CRC press, 2018. [32] Ehsan Toreini, Mhairi Aitken, Kovila Coopamootoo, Karen Elliott, Carlos Gonzalez Zelaya, and Aad van Moorsel. The relationship between trust in AI and trustworthy machine learning technologies. In Proceedings of the 2020 Conference on Fairness, Accountability, and Transparency, pages 272–283, 2020. [33] Michael Veale and Reuben Binns. Fairer machine learning in the real world: Mitigating discrimination without collecting sensitive data. Big Data & Society, 4(2):2053951717743530, 2017. [34] Tong Wang, Cynthia Rudin, Finale Doshi-Velez, Yimin Liu, Erica Klampfl, and Perry MacNeille. A bayesian framework for learning rule sets for interpretable classification. The Journal of Machine Learning Research, 18(1):2357–2393, 2017. [35] Qiang Yang, Yang Liu, Tianjian Chen, and Yongxin Tong. Federated machine learning: Concept and applications. ACM Transactions on Intelligent Systems and Technology (TIST), 10(2):1–19, 2019.

Privacy-preserving Computation of Fairness for ML Systems: Acknowledgement & References

This paper is under CC BY 4.0 DEED license. available on arxiv Authors: (1) Ehsan Toreini, University of Surrey, UK; (2) Maryam Mehrnezhad, Royal Holloway University of London; (3) Aad Van Moorsel, Birmingham University. Table of Links Abstract & Introduction Background and Related Work FaaS Architecture Implementation and Performance Analysis Conclusion Acknowledgment and References 5 Conclusion This paper proposes Fairness as a Service (FaaS), a trustworthy service architecture and secure protocol for the calculation of algorithmic fairness. FaaS is designed as a service that calculates fairness without asking the ML system to share the original dataset or model information. Instead, it requires an encrypted representation of the values of the data features delivered by the ML system in the shape of cryptograms. We used non-interactive Zero Knowledge Proofs within the cryptogram to assure that the protocol is executed as it should. These cryptograms are posted on a public fairness board for everyone to inspect the correctness of the computations for the fairness of the ML system. This is a new approach in privacy–preserving computation of fairness since unlike other similar proposals that use federated learning approach, our FaaS architecture does not rely on a specific machine learning model or a fairness metric definition for its operation. Instead, one have the freedom of deploying their desired model and the fairness metric of choice. In this paper we proved that the security protocol guarantees the privacy of data and does not leak any model information. Compared to earlier designs, trust in our design is in the correct construction of the cryptogram by the ML system. Arguably, this is more realistic as a solution than providing full access to data to the trusted third party, taking into account the many legal, business and ethical requirements of ML systems. At the same time, this provides a new challenge in increasing the trust one has in the ML system. Increasing trust in the construction of the cryptograms remains an interesting research challenge following from the presented protocol. We implemented a proof-of-concept of FaaS and conducted performance experiments on commodity hardware. The protocol takes seconds per data point to complete, thus demonstrating in performance challenges if the number of data points is large (tens of thousands). To mitigate the performance challenge, the security protocol is staged such that the construction of the cryptogram can be done off-line. The performance of the calculation of fairness from the cryptogram is a challenge to address in future work. All together, we believe FaaS and the presented underlying security protocol provide a new and promising approach to calculating and verifying fairness of AI algorithms.

COMPSTAK

compstak.com

51-200 employees

Since 2011

COMPANY RANKING

COMPSTAK

EVERGREEN INDEX #2821

Representing this company? Click here to claim and customize this page!

CompStak WIKI

Apply to get your Tech Company News Page live on HackerNoon.com today!

COMPSTAK

compstak.com

51-200 employees

Since 2011

COMPANY RANKING

COMPSTAK

EVERGREEN INDEX #2821

RELATED COMPANIESRANK

STORIES ABOUT COMPSTAK

Representing this company? Click here to claim and customize this page!

CompStak WIKI

Apply to get your Tech Company News Page live on HackerNoon.com today!