Search

Outlier detection is critical to a large number of applications from finance fraud detection to health care. Although numerous approaches have been proposed to automatically detect outliers, such outliers detected based on statistical rarity do not necessarily correspond to the true outliers to the interest of applications. In this work, we propose a human-in-the-loop outlier detection approach HOD that effectively leverages human intelligence to discover the true outliers. There are two main challenges in HOD. The first is to design human-friendly questions such that humans can easily understand the questions even if humans know nothing about the outlier detection techniques. The second is to minimize the number of questions. To address the first challenge, we design a clustering-based method to effectively discover a small number of objects that are unlikely to be outliers (aka, inliers) and yet effectively represent the typical characteristics of the given dataset. HOD then leverages this set of inliers (called context inliers) to help humans understand the context in which the outliers occur. This ensures humans are able to easily identify the true outliers from the outlier candidates produced by the machine-based outlier detection techniques. To address the second challenge, we propose a bipartite graph-based question selection strategy that is theoretically proven to be able to minimize the number of questions needed to cover all outlier candidates. Our experimental results on real data sets show that HOD significantly outperforms the state-of-the-art methods on both human efforts and the quality of the discovered outliers.

Exploration in novel settings can be challenging without prior experience in similar domains. However, humans are able to build on prior experience quickly and efficiently. Children exhibit this behavior when playing with toys. For example, given a toy with a yellow and blue door, a child will explore with no clear objective, but once they have discovered how to open the yellow door, they will most likely be able to open the blue door much faster. Adults also exhibit this behavior when entering new spaces such as kitchens. We develop a method, Contextual Prior Prediction, which provides a means of transferring knowledge between interactions in similar domains through vision. We develop agents that exhibit exploratory behavior with increasing efficiency, by learning visual features that are shared across environments, and how they correlate to actions. Our problem is formulated as a Contextual Multi-Armed Bandit where the contexts are images, and the robot has access to a parameterized action space. Given a novel object, the objective is to maximize reward with few interactions. A domain which strongly exhibits correlations between visual features and motion is kinemetically constrained mechanisms. We evaluate our method on simulated prismatic and revolute joints., Comment: IEEE International Conference on Robotics and Automation (ICRA) 2020

Deep exploration requires coordinated long-term planning. We present a model-based reinforcement learning algorithm that guides policy learning through a value function that exhibits optimism in the face of uncertainty. We capture uncertainty over values by combining predictions from an ensemble of models and formulate an upper confidence bound (UCB) objective to recover optimistic estimates. Training the policy on ensemble rollouts with the learned value function as the terminal cost allows for projecting long-term interactions into a limited planning horizon, thus enabling deep optimistic exploration. We do not assume a priori knowledge of either the dynamics or reward function. We demonstrate that our approach can accommodate both dense and sparse reward signals, while improving sample complexity on a variety of benchmarking tasks. Keywords: Reinforcement Learning; Deep Exploration; Model-Based; Value Function; UCB, Office of Naval Research; Qualcomm; Toyota Research Institute

Similarity search is the basis for many data analytics techniques, including k-nearest neighbor classification and outlier detection. Similarity search over large data sets relies on i) a distance metric learned from input examples and ii) an index to speed up search based on the learned distance metric. In interactive systems, input to guide the learning of the distance metric may be provided over time. As this new input changes the learned distance metric, a naive approach would adopt the costly process of re-indexing all items after each metric change. In this paper, we propose the first solution, called OASIS, to instantaneously adapt the index to conform to a changing distance metric without this prohibitive re-indexing process. To achieve this, we prove that locality-sensitive hashing (LSH) provides an invariance property, meaning that an LSH index built on the original distance metric is equally effective at supporting similarity search using an updated distance metric as long as the transform matrix learned for the new distance metric satisfies certain properties. This observation allows OASIS to avoid recomputing the index from scratch in most cases. Further, for the rare cases when an adaption of the LSH index is shown to be necessary, we design an efficient incremental LSH update strategy that re-hashes only a small subset of the items in the index. In addition, we develop an efficient distance metric learning strategy that incrementally learns the new metric as inputs are received. Our experimental study using real world public datasets confirms the effectiveness of OASIS at improving the accuracy of various similarity search-based data analytics tasks by instantaneously adapting the distance metric and its associated index in tandem, while achieving an up to 3 orders of magnitude speedup over the state-of-art techniques.

Background: A digital version of the clock drawing test (dCDT) provides new latency and graphomotor behavioral measurements. These variables have yet to be validated with external neuropsychological domains in non-demented adults. Objective: The current investigation reports on cognitive constructs associated with selected dCDT latency and graphomotor variables and compares performances between individuals with mild cognitive impairment (MCI) and non-MCI peers. Methods: 202 non-demented older adults (age 68.79 ± 6.18, 46% female, education years 16.02 ± 2.70) completed the dCDT and a comprehensive neuropsychological protocol. dCDT variables of interest included: total completion time (TCT), pre-first hand latency (PFHL), post-clock face latency (PCFL), and clock face area (CFA). We also explored variables of percent time drawing (i.e., 'ink time') versus percent time not drawing (i.e., 'think time'). Neuropsychological domains of interest included processing speed, working memory, language, and declarative memory. Results: Adjusting for age and premorbid cognitive reserve metrics, command TCT positively correlated with multiple cognitive domains; PFHL and PCFL negatively associated with worse performance on working memory and processing speed tests. For Copy, TCT, PCFL, and PFHL negatively correlated with processing speed, and CFA negatively correlated with language. Between-group analyses show MCI participants generated slower command TCT, produced smaller CFA, and required more command 'think' (% Think) than 'ink' (% Ink) time. Conclusion: Command dCDT variables of interest were primarily processing speed and working memory dependent. MCI participants showed dCDT differences relative to non-MCI peers, suggesting the dCDT may assist with classification. Results document cognitive construct validation to digital metrics of clock drawing., National Science Foundation (Grant 1404333)

The resurging interest in Byzantine fault tolerant systems will demand more scalable threshold cryptosystems. Unfortunately, current systems scale poorly, requiring time quadratic in the number of participants. In this paper, we present techniques that help scale threshold signature schemes (TSS), verifiable secret sharing (VSS) and distributed key generation (DKG) protocols to hundreds of thousands of participants and beyond. First, we use efficient algorithms for evaluating polynomials at multiple points to speed up computing Lagrange coefficients when aggregating threshold signatures. As a result, we can aggregate a 130,000 out of 260,000 BLS threshold signature in just 6 seconds (down from 30 minutes). Second, we show how "authenticating" such multipoint evaluations can speed up proving polynomial evaluations, a key step in communication-efficient VSS and DKG protocols. As a result, we reduce the asymptotic (and concrete) computational complexity of VSS and DKG protocols from quadratic time to quasilinear time, at a small increase in communication complexity. For example, using our DKG protocol, we can securely generate a key for the BLS scheme above in 2.3 hours (down from 8 days). Our techniques improve performance for thresholds as small as 255 and generalize to any Lagrange-based threshold scheme, not just threshold signatures. Our work has certain limitations: we require a trusted setup, we focus on synchronous VSS and DKG protocols and we do not address the worst-case complaint overhead in DKGs. Nonetheless, we hope it will spark new interest in designing large-scale distributed systems.

Users have long struggled to extract and repurpose data from websites by laboriously copying or scraping content from web pages. An alternative is to write scripts that pull data through APIs. This provides a cleaner way to access data than scraping; however, APIs are effortful for programmers and nigh-impossible for non-programmers to use. In this work, we empower users to access APIs without programming. We evolve a schema for declaratively specifying how to interact with a data API. We then develop ScrAPIr: a standard query GUI that enables users to fetch data through any API for which a specification exists, and a second GUI that lets users author and share the specification for a given API. From a lab evaluation, we find that even non-programmers can access APIs using ScrAPIr, while programmers can access APIs 3.8 times faster on average using ScrAPIr than using programming.

We introduce SparseVM, a method that registers clinical-quality 3D MR scans both faster and more accurately than previously possible. Deformable alignment, or registration, of clinical scans is a fundamental task for many clinical neuroscience studies. However, most registration algorithms are designed for high-resolution research-quality scans. In contrast to research-quality scans, clinical scans are often sparse, missing up to 86% of the slices available in research-quality scans. Existing methods for registering these sparse images are either inaccurate or extremely slow. We present a learning-based registration method, SparseVM, that is more accurate and orders of magnitude faster than the most accurate clinical registration methods. To our knowledge, it is the first method to use deep learning specifically tailored to registering clinical images. We demonstrate our method on a clinically-acquired MRI dataset of stroke patients and on a simulated sparse MRI dataset. Our code is available as part of the VoxelMorph package at http://voxelmorph.mit.edu/., Comment: This version was accepted to CHIL. It builds on the previous version of the paper and includes more experimental results

© 2020 ACM. We present Sprayable User Interfaces: room-sized interactive surfaces that contain sensor and display elements created by airbrushing functional inks. Since airbrushing is inherently mobile, designers can create large-scale user interfaces on complex 3D geometries where existing stationary fabrication methods fail. To enable Sprayable User Interfaces, we developed a novel design and fabrication pipeline that takes a desired user interface layout as input and automatically generates stencils for airbrushing the layout onto a physical surface. After fabricating stencils from cardboard or projecting stencils digitally, designers spray each layer with an airbrush, attach a microcontroller to the user interface, and the interface is ready to be used. Our technical evaluation shows that Sprayable User Interfaces work on various geometries and surface materials, such as porous stone and rough wood. We demonstrate our system with several application examples including interactive smart home applications on a wall and a soft leather sofa, an interactive smart city application, and interactive architecture in public office spaces.

Communication is critical to collaboration; however, too much of it can degrade performance. Motivated by the need for effective use of a robot’s communication modalities, in this work, we present a computational framework that decides if, when, and what to communicate during human-robot collaboration. The framework, titled CommPlan, consists of a model specification process and an execution-time POMDP planner. To address the challenge of collecting interaction data, the model specification process is hybrid: where part of the model is learned from data, while the remainder is manually specified. Given the model, the robot’s decision-making is performed computationally during interaction and under partial observability of human’s mental states. We implement CommPlan for a shared workspace task, in which the robot has multiple communication options and needs to reason within a short time. Through experiments with human participants, we confirm that CommPlan results in the effective use of communication capabilities and improves human-robot collaboration. ACM Reference Format: Vaibhav V. Unhelkar, Shen Li, and Julie A. Shah. 2020. Decision-Making for Bidirectional Communication in Sequential Human-Robot Collaborative Tasks. In Proceedings of the 2020 ACM/IEEE International Conference on Human-Robot Interaction (HRI ’20), March 23–26, 2020, Cambridge, United Kingdom. ACM, New York, NY, USA, 13 pages. https://doi.org/10.1145/3319502.3374779

The rapid adoption of Reconfigurable Optical Add-Drop Multiplexers (ROADMs) is setting the stage for the dynamic reconfiguration of the network topology in optical backbones. The conventional approach to enable programmability in the physical layer requires solving a cross-layer optimization formulation that captures the interplay between the IP and optical layers. However, as the network scales, the complexity and run time of cross-layer optimization formulations grow prohibitively, resulting in heuristic-based solutions that sacrifice optimality for scalability. We propose a flow-based graph abstraction, called OptFlow, that is able to find the optimal allocation faster than a cross-layer optimization formulation. The key idea in OptFlow is that topology programmability is abstracted by "network flows," enabling service providers to use multi-commodity flow formulations, such as conventional Traffic Engineering, to solve a cross-layer optimization. OptFlow augments the physical graph and uses it as input to the unmodified flow-based Traffic Engineering algorithm, capturing a variety of IP-layer optimization goals such as max throughput, min hop count, and max-min fairness. Due to its flow-based nature, OptFlowinherently provides an abstraction for consistent network updates. To benchmark our key assumptions in OptFlow, we build a small testbed prototype consisting of four ROADMs. To evaluate the optimality and run time of large networks, we simulate fiveWAN topologies with up to 100 nodes and 390 links. Our results show that OptFlow matches the throughput performance of an optimal cross-layer formulation but has faster computation times. The run time speed-up of OptFlow increases as the network scales, with up to 8× faster execution times in our simulations., European Union. Horizon 2020 Research and Innovation Programme (Agreement 864228)

© 2019 IEEE. In this paper, we address the problem of closed-loop control of nonlinear dynamical systems subjected to probabilistic uncertainties. More precisely, we design time-varying polynomial feedback controllers to follow the given nominal trajectory and also, for safety purposes, remain in the tube around the nominal trajectory, despite all uncertainties. We formulate this problem as a chance optimization problem where we maximize the probability of achieving control objectives. To address control problems with long planning horizons, we formulate the single large chance optimization problem as a sequence of smaller chance optimization problems. To solve the obtained chance optimization problems, we leverage the theory of measures and moments and obtain convex relaxations in the form of semidefinite programs. We provide numerical examples on stabilizing controller design and motion planning of uncertain nonlinear systems to illustrate the performance of the proposed approach.

© 2020 ACM. Imagine a future where a domestic robot ships with a state-of-theart learning from demonstrations (LfD) system to learn household tasks. You would like the robot to set the dinner-table for you when you get home at dinner time. After you demonstrate how to set the dinner table a couple of times. Would you be confident that robot will not try to place the saucer on top of the cup, or finish as much of the task as possible if an object was missing?

We present a method and software for ballot-polling risk-limiting audits (RLAs) based on Bernoulli sampling: ballots are included in the sample with probability p, independently. Bernoulli sampling has several advantages: (1) it does not require a ballot manifest; (2) it can be conducted independently at different locations, rather than requiring a central authority to select the sample from the whole population of cast ballots or requiring stratified sampling; (3) it can start in polling places on election night, before margins are known. If the reported margins for the 2016 U.S. Presidential election are correct, a Bernoulli ballot-polling audit with a risk limit of 5% and a sampling rate of p0=1% would have had at least a 99% probability of confirming the outcome in 42 states. (The other states were more likely to have needed to examine additional ballots). Logistical and security advantages that auditing in the polling place affords may outweigh the cost of examining more ballots than some other methods might require.

Reconfigurable analog devices are a powerful new computing substrate especially appropriate for executing computationally intensive dynamical system computations in an energy efficient manner. We present Legno, a compilation toolchain for programmable analog devices. Legno targets the HCDCv2, a programmable analog device designed to execute general nonlinear dynamical systems. To the best of our knowledge, Legno is the first compiler to successfully target a physical (as opposed to simulated) programmable analog device for dynamical systems and this paper is the first to present experimental results for any compiled computation executing on any physical programmable analog device of this class. The Legno compiler synthesizes analog circuits from parametric and specialized blocks and accounts for analog noise, quantization error, and manufacturing variations within the device. We evaluate the compiled configurations on the Sendyne S100Asy RevU development board on twelve benchmarks from physics, controls, and biology. Our results show that Legno produces accurate computations on the analog device. The computations execute in 0.50-5.92 ms and consume 0.28-5.67 µJ of energy., United States. Defense Advanced Research Projects Agency (Grants HACCS HR001118C0059 and TC FA8650-15-C-7564)

Due to the rapid emergence of antibiotic-resistant bacteria, there is a growing need to discover new antibiotics. To address this challenge, we trained a deep neural network capable of predicting molecules with antibacterial activity. We performed predictions on multiple chemical libraries and discovered a molecule from the Drug Repurposing Hub—halicin—that is structurally divergent from conventional antibiotics and displays bactericidal activity against a wide phylogenetic spectrum of pathogens including Mycobacterium tuberculosis and carbapenem-resistant Enterobacteriaceae. Halicin also effectively treated Clostridioides difficile and pan-resistant Acinetobacter baumannii infections in murine models. Additionally, from a discrete set of 23 empirically tested predictions from >107 million molecules curated from the ZINC15 database, our model identified eight antibacterial compounds that are structurally distant from known antibiotics. This work highlights the utility of deep learning approaches to expand our antibiotic arsenal through the discovery of structurally distinct antibacterial molecules. A trained deep neural network predicts antibiotic activity in molecules that are structurally different from known antibiotics, among which Halicin exhibits efficacy against broad-spectrum bacterial infections in mice., Defence Threat Reduction Agency (Grant HDTRA1-15- 1-0051)

© 2020 Association for Computing Machinery. We present NURBSforms: A modular shape-changing interface for prototyping curved surfaces. Each NURBSform module represents an edge of variable curvature that, when joined together with other modules, enables designers to construct surfaces and adjust their curvature interactively. NURBSform modules vary their curvature using active and passive shape memory materials: An embedded shape memory alloy (SMA) wire increases the curvature when heated, while an elastic material recovers the flat shape when the SMA wire cools down. A hall effect sensor on each module allows users to vary the curvature by adjusting the distance of their hand. In addition, NURBSforms provides functions across multiple modules, such as ?save', ?reset', and ?load', to facilitate design exploration. Since each module is self-contained and individually controllable, NURBSform modules scale well and can be connected into large networks of curves representing various geometries. By giving examples of different NURBSforms assemblies, we demonstrate how the modularity of NURBSforms, together with its integrated computational support, enables designers to quickly explore different versions of a shape in a single integrated design process.

New consent management platforms (CMPs) have been introduced to the web to conform with the EU's General Data Protection Regulation, particularly its requirements for consent when companies collect and process users' personal data. This work analyses how the most prevalent CMP designs affect people's consent choices. We scraped the designs of the five most popular CMPs on the top 10,000 websites in the UK (n=680). We found that dark patterns and implied consent are ubiquitous; only 11.8% meet the minimal requirements that we set based on European law. Second, we conducted a field experiment with 40 participants to investigate how the eight most common designs affect consent choices. We found that notification style (banner or barrier) has no effect; removing the opt-out button from the first page increases consent by 22--23 percentage points; and providing more granular controls on the first page decreases consent by 8--20 percentage points. This study provides an empirical basis for the necessary regulatory action to enforce the GDPR, in particular the possibility of focusing on the centralised, third-party CMP services as an effective way to increase compliance., 13 pages, 3 figures. To appear in the Proceedings of CHI '20 CHI Conference on Human Factors in Computing Systems, April 25--30, 2020, Honolulu, HI, USA

During the years, it has been possible to assess significant improvements in the computational efficiency of Semantic Web search and exploration systems. However, it has been much harder to assess how well different semantic systems’ user interfaces help their users. One of the key factors facilitating the advancement of research in a particular field is the ability to compare the performance of different approaches. Though there are many such benchmarks in Semantic Web fields that have experienced significant improvements, this is not the case for Semantic Web user interfaces for data exploration. We propose and demonstrate the use of a benchmark for evaluating such user interfaces, which includes a set of typical user tasks and a well-defined procedure for assigning a measure of performance on those tasks to a semantic system. We have applied the benchmark to four such systems. Moreover, all the required resources to apply the benchmark are openly available online. We intend to initiate a community conversation that will lead to a generally accepted framework for comparing systems and for measuring, and thus encouraging, progress towards better semantic search and exploration tools.

We design a Local Computation Algorithm (LCA) for the set cover problem. Given a set system where each set has size at most s and each element is contained in at most t sets, the algorithm reports whether a given set is in some fixed set cover whose expected size is O(log s) times the minimum fractional set cover value. Our algorithm requires s (log s t (log s·(log log s+log log t)) queries. This result improves upon the application of the reduction of [Parnas and Ron, TCS'07] on the result of [Kuhn et al., SODA'06], which leads to a query complexity of (st) (log s·log t . To obtain this result, we design a parallel set cover algorithm that admits an efficient simulation in the LCA model by using a sparsification technique introduced in [Ghaffari and Uitto, SODA'19] for the maximal independent set problem. The parallel algorithm adds a random subset of the sets to the solution in a style similar to the PRAM algorithm of [Berger et al., FOCS'89]. However, our algorithm differs in the way that it never revokes its decisions, which results in a fewer number of adaptive rounds. This requires a novel approximation analysis which might be of independent interest. O ) O O ), Swiss National Science Foundation (Grant P2ELP2_181772), MIT-IBM Watson AI Lab (Rsearch Collaboration Agreement W1771646), National Science Foundation (U.S.) (Awards CCF-1740751, CCF-1733808, and IIS-1741137)

This paper presents an efficient approach to quantifying image registration uncertainty based on a low-dimensional representation of geometric deformations. In contrast to previous methods, we develop a Bayesian diffeomorphic registration framework in a bandlimited space, rather than a high-dimensional image space. We show that a dense posterior distribution on deformation fields can be fully characterized by much fewer parameters, which dramatically reduces the computational complexity of model inferences. To further avoid heavy computation loads introduced by random sampling algorithms, we approximate a marginal posterior by using Laplace’s method at the optimal solution of log-posterior distribution. Experimental results on both 2D synthetic data and real 3D brain magnetic resonance imaging (MRI) scans demonstrate that our method is significantly faster than the state-of-the-art diffeomorphic registration uncertainty quantification algorithms, while producing comparable results.

The polytope containment problem is deciding whether a polytope is a contained within another polytope. This problem is rooted in computational convexity, and arises in applications such as verification and control of dynamical systems. The complexity heavily depends on how the polytopes are represented. Describing polytopes by their hyperplanes (H-polytopes) is a popular representation. In many applications we use affine transformations of H-polytopes, which we refer to as AH-polytopes. Zonotopes, orthogonal projections of H-polytopes, and convex hulls/Minkowski sums of multiple H-polytopes can be efficiently represented as AH-polytopes. While there exists efficient necessary and sufficient conditions for AH-polytope in H-polytope containment, the case of AH-polytope in AH-polytope is known to be NP-complete. In this paper, we provide a sufficient condition for this problem that is cast as a linear program with size that grows linearly with the number of hyperplanes of each polytope. Special cases on zonotopes, Minkowski sums, convex hulls, and disjunctions of H-polytopes are studied. These efficient encodings enable us to designate certain components of polytopes as decision variables, and incorporate them into a convex optimization problem. We present examples on the zonotope containment problem, polytopic Hausdorff distances, zonotope order reduction, inner approximations of orthogonal projections, and demonstrate the usefulness of our results on formal controller verification and synthesis for hybrid systems., Comment: Shorter version submitted to CDC 2019

Conventional buffer sizing techniques consider an output port with multiple queues in isolation and provide guidelines for the size of the queue. In practice, however, switches consist of several ports that share a buffering chip. Hence, chip manufacturers, such as Broadcom, are left to devise a set of proprietary resource sharing algorithms to allocate buffers across ports. This algorithm dynamically adjusts the buffer size for output queues and directly impacts the packet loss and latency of individual queues. We show that the problem of allocating buffers across ports, although less known, is indeed responsible for fundamental inefficiencies in today's devices. In particular, the per-port buffer allocation is an ad-hoc decision that (at best) depends on the remaining buffer cells on the chip instead of the type of traffic. In this work, we advocate for a flow-aware and device-wide buffer sharing scheme (FAB), which is practical today in programmable devices. We tested FAB on two specific workloads and showed that it can improve the tail flow completion time by an order of magnitude compared to conventional buffer management techniques.

We present ClearTrack, a system that tracks 32 bits of metadata for each primitive value in Java programs to detect and nullify a range of vulnerabilities such as integer overflow and underflow vulnerabilities, SQL injection vulnerabilities, and command injection vulnerabilities. Contributions include new techniques for eliminating false positives associated with benign integer overflows and underflows, new metadata-aware techniques for detecting and nullifying SQL and command injection attacks, and results from an evaluation of ClearTrack performed by a Test and Evaluation team hired by the sponsor of this research (an anonymous agency of the United States government). These results show that 1) ClearTrack operates successfully on Java programs comprising hundreds of thousands of lines of code (including instrumented jar files and Java system libraries, the majority of the applications comprise over 3 million lines of code), 2) because of computations such as cryptography and hash table calculations, these applications perform millions of benign integer overflows and underflows, and 3) ClearTrack successfully detects and nullifies all tested integer overflow and underflow, SQL injection, and command injection vulnerabilities in the benchmark applications.

In the noisy tensor completion problem we observe $m$ entries (whose location is chosen uniformly at random) from an unknown $n_1 \times n_2 \times n_3$ tensor $T$. We assume that $T$ is entry-wise close to being rank $r$. Our goal is to fill in its missing entries using as few observations as possible. Let $n = \max(n_1, n_2, n_3)$. We show that if $m = n^{3/2} r$ then there is a polynomial time algorithm based on the sixth level of the sum-of-squares hierarchy for completing it. Our estimate agrees with almost all of $T$'s entries almost exactly and works even when our observations are corrupted by noise. This is also the first algorithm for tensor completion that works in the overcomplete case when $r > n$, and in fact it works all the way up to $r = n^{3/2-\epsilon}$. Our proofs are short and simple and are based on establishing a new connection between noisy tensor completion (through the language of Rademacher complexity) and the task of refuting random constant satisfaction problems. This connection seems to have gone unnoticed even in the context of matrix completion. Furthermore, we use this connection to show matching lower bounds. Our main technical result is in characterizing the Rademacher complexity of the sequence of norms that arise in the sum-of-squares relaxations to the tensor nuclear norm. These results point to an interesting new direction: Can we explore computational vs. sample complexity tradeoffs through the sum-of-squares hierarchy?, Comment: 24 pages

As bytes-per-FLOP ratios continue to decline, communication is becoming a bottleneck for performance scaling. This paper describes bandwidth steering in HPC using emerging reconfigurable silicon photonic switches. We demonstrate that placing photonics in the lower layers of a hierarchical topology efficiently changes the connectivity and consequently allows operators to recover from system fragmentation that is otherwise hard to mitigate using common task placement strategies. Bandwidth steering enables efficient utilization of the higher layers of the topology and reduces cost with no performance penalties. In our simulations with a few thousand network endpoints, bandwidth steering reduces static power consumption per unit throughput by 36% and dynamic power consumption by 14% compared to a reference fat tree topology. Such improvements magnify as we taper the bandwidth of the upper network layer. In our hardware testbed, bandwidth steering improves total application execution time by 69%, unaffected by bandwidth tapering., United States. Advanced Research Projects Agency-Energy. ENLITENED Program (Project award DE-AR00000843), United States. Department of Energy. Office of Science ( Contract DE-AC02-05CH11231)

Current methods for autonomous management use strict first-come, first-serve (FCFS) ordering to manage incoming autonomous vehicles at an intersection. In this work, we present a coordination policy that swaps agent ordering to increase the system-wide performance while ensuring that the swaps are socially compliant. By considering an agent’s Social Value Orientation (SVO), a social psychology metric for their willingness to help another vehicle, the central coordinator can reduce system delays while ensuring each individual vehicle increases their own utility. The FCFS-SVO algorithm is both computationally tractable and accounts for a variety of real-world agent types, such as human drivers and a variety of social orientations. Simulation results show that average vehicle delays decrease with swapping by enabling cooperation between agents. In addition, we show that the proportion of human drivers, as well as, the distribution of prosocial and egoistic vehicles in the system can have a prominent effect on the performance of the system.

We construct private-key and public-key functional encryption schemes in the bounded-key setting; that is, secure against adversaries that obtain an a-priori bounded number of functional keys (also known as the collusion bound). An important metric considered in the literature on bounded-key functional encryption schemes is the dependence of the running time of the encryption algorithm on the collusion bound Q = Q(λ) (where λ is the security parameter). It is known that bounded-key functional encryption schemes with encryption complexity growing with ε > 0, for any constant Q1-λ, implies indistinguishability obfuscation. On the other hand, in the public-key setting, it was previously unknown whether we could achieve encryption complexity growing linear with Q, also known as optimal bounded-key FE, based on well-studied assumptions. In this work, we give the first construction of an optimal bounded-key public-key functional encryption scheme under the minimal assumption of the existence of any public-key encryption scheme. Moreover, our scheme supports the class of all polynomial-size circuits. Our techniques also extend to the private-key setting. We achieve a construction of an optimal bounded-key functional encryption in the private-key setting based on the minimal assumption of one-way functions, instead of learning with errors as achieved in prior works.

Compilers and performance engineers use hardware performance models to simplify program optimizations. Performance models provide a necessary abstraction over complex modern processors. However, constructing and maintaining a performance model can be onerous, given the numerous microarchi-tectural optimizations employed by modern processors. Despite their complexity and reported inaccuracy (e.g., deviating from native measurement by more than 30%), existing performance models-such as IACA and llvm-mca-have not been systematically validated, because there is no scalable machine code profiler that can automatically obtain throughput of arbitrary basic blocks while conforming to common modeling assumptions. In this paper, we present a novel profiler that can profile arbitrary memory-accessing basic blocks without any user intervention. We used this profiler to build BHive, a benchmark for systematic validation of performance models of x86-64 basic blocks. We used BHive to evaluate four existing performance models: IACA, llvm-mca, Ithemal, and OSACA. We automatically cluster basic blocks in the benchmark suite based on their utilization of CPU resources. Using this clustering, our benchmark can give a detailed analysis of a performance model's strengths and weaknesses on different workloads (e.g., vectorized vs. scalar basic blocks). We additionally demonstrate that our dataset well captures basic properties of two Google applications: Spanner and Dremel.

Proton beam therapy is the current state of the art for irradiation of cancerous tumors near vital organs. Proton imaging is a new technology that images tumors with the same particle interaction as proton therapy, promising more accurate treatment preparation compared to traditional computerized tomography scans. A specific proton imaging technique in development at Massachusetts General Hospital requires an energy modulator that can fit inside the gantry nozzle. This work presents the design of a compact proton beam energy modulator required for such imaging applications. A combination of steel and lead wedges was used to create a balanced modulation wheel, with an axis of rotation perpendicular to the incident beam. The mechanical design also allows interchangeable disks and precise position control. The modulator design was verified by testing on a proton beam line. The final device achieved a wide range of energy modulation (21 cm water-equivalent thickness) while maintaining a constant exiting beam angular spread meeting device requirements. This compact design facilitates proton imaging to be practically incorporated into future gantry systems, which will advance proton treatment for tumors near vital organs.

© 2019 IEEE. We present a trajectory planning algorithm for the shapeshifting of reconfigurable modular surface vessels. Each vessel is designed to latch with and unlatch from other vessels, which we aim to use to create dynamic infrastructure, such as on-demand bridges and temporary market squares, in canal environments. Our algorithm generates smooth and collision-free trajectories that the vessels can track to reconfigure their connections. We formulate the trajectory planning problem as Mixed Integer Quadratic Programming (MIQP) with a B-spline representation. We conceive a physical platform of the reconfigurable modular vessels and, through swimming pool experiments, show the efficacy of our trajectory planning algorithm for the shapeshifting of the vessels.

© 2019, International Association for Cryptologic Research. Middle-product learning with errors (MP-LWE) was recently introduced by Rosca, Sakzad, Steinfeld and Stehlé (CRYPTO 2017) as a way to combine the efficiency of Ring-LWE with the more robust security guarantees of plain LWE. While Ring-LWE is at the heart of efficient lattice-based cryptosystems, it involves the choice of an underlying ring which is essentially arbitrary. In other words, the effect of this choice on the security of Ring-LWE is poorly understood. On the other hand, Rosca et al. showed that a new LWE variant, called MP-LWE, is as secure as Polynomial-LWE (another variant of Ring-LWE) over any of a broad class of number fields. They also demonstrated the usefulness of MP-LWE by constructing an MP-LWE based public-key encryption scheme whose efficiency is comparable to Ring-LWE based public-key encryption. In this work, we take this line of research further by showing how to construct Identity-Based Encryption (IBE) schemes that are secure under a variant of the MP-LWE assumption. Our IBE schemes match the efficiency of Ring-LWE based IBE, including a scheme in the random oracle model with keys and ciphertexts of size (formula presented) (for n-bit identities). We construct our IBE scheme following the lattice trapdoors paradigm of [Gentry, Peikert, and Vaikuntanathan, STOC’08]; our main technical contributions are introducing a new leftover hash lemma and instantiating a new variant of lattice trapdoors compatible with MP-LWE. This work demonstrates that the efficiency/security tradeoff gains of MP-LWE can be extended beyond public-key encryption to more complex lattice-based primitives.

© 2019 IEEE. As robots and other autonomous agents are increasingly incorporated into complex domains, characterizing interaction within heterogeneous teams that include both humans and machines becomes more necessary. Previous literature has addressed the task of characterizing human-robot interaction from different perspectives and in multiple contexts. However, the numerous factors behind interaction work in conjunction, and the insights gained from one perspective can inadvertently affect another, creating a need for unification of these taxonomies and frameworks within an overarching taxonomy that systematically defines these relationships. In this paper we review existing taxonomies related to human-robot interaction, the behavioral sciences, and social and algorithmic taxonomies, and propose an overarching ontology for the factors from these works. We identify three main components characterizing the structure of an interaction (environment, task, and team), and structure them over two levels: contextual factors and factors driven by local dynamics. Finally, we present an analysis of how these factors affect decisions about levels of robot automation and level of information abstraction in an interaction, and discuss curent gaps in the literature that can motivate future research.

Current perception systems mostly require direct line of sight to anticipate and ultimately prevent potential collisions at intersections with other road users. We present a fully integrated autonomous system capable of detecting shadows or weak illumination changes on the ground caused by a dynamic obstacle in NLoS scenarios. This additional virtual sensor “ShadowCam” extends the signal range utilized so far by computer-vision ADASs. We show that (1) our algorithm maintains the mean classification accuracy of around 70% even when it doesn’t rely on infrastructure – such as AprilTags - as an image registration method. We validate (2) in real-world experiments that our autonomous car driving in night time conditions detects a hidden approaching car earlier with our virtual sensor than with the front facing 2-D LiDAR.

Notary is a new hardware and software architecture for running isolated approval agents in the form factor of a USB stick with a small display and buttons. Approval agents allow factoring out critical security decisions, such as getting the user’s approval to sign a Bitcoin transaction or to delete a backup, to a secure environment. The key challenge addressed by Notary is to securely switch between agents on the same device. Prior systems either avoid the problem by building single-function devices like a USB U2F key, or they provide weak isolation that is susceptible to kernel bugs, side channels, or Rowhammer-like attacks. Notary achieves strong isolation using reset-based switching, along with the use of physically separate systems-on-a-chip for agent code and for the kernel, and a machine-checked proof of both the hardware’s register-transfer-level design and software, showing that reset-based switching leaks no state. Notary also provides a trustworthy I/O path between the agent code and the user, which prevents an adversary from tampering with the user’s screen or buttons. We built a hardware/software prototype of Notary, using a combination of ARM and RISC-V processors. The prototype demonstrates that it is feasible to verify Notary’s reset-based switching, and that Notary can support diverse agents, including cryptocurrencies and a transaction approval agent for traditional client-server applications such as websites. Measurements of reset-based switching show that it is fast enough for interactive use. We analyze security bugs in existing cryptocurrency hardware wallets, which aim to provide a similar form factor and feature set as Notary, and show that Notary’s design avoids many bugs that affect them., NSF (Awards CNS-1413920 and CNS-1812522)

When can a polyomino piece of paper be folded into a unit cube? Prior work studied tree-like polyominoes, but polyominoes with holes remain an intriguing open problem. We present sufficient conditions for a polyomino with one or several holes to fold into a cube, and conditions under which cube folding is impossible. In particular, we show that all but five special \emph{simple} holes guarantee foldability., 24 pages, 21 figures

Collections of images under a single, uncontrolled illumination have enabled the rapid advancement of core computer vision tasks like classification, detection, and segmentation. But even with modern learning techniques, many inverse problems involving lighting and material understanding remain too severely ill-posed to be solved with single-illumination datasets. The data simply does not contain the necessary supervisory signals. Multi-illumination datasets are notoriously hard to capture, so the data is typically collected at small scale, in controlled environments, either using multiple light sources, or robotic gantries. This leads to image collections that are not representative of the variety and complexity of real world scenes. We introduce a new multi-illumination dataset of more than 1000 real scenes, each captured in high dynamic range and high resolution, under 25 lighting conditions. We demonstrate the richness of this dataset by training state-of-the-art models for three challenging applications: Single-image illumination estimation, image relighting, and mixed-illuminant white balance., United States. Defense Advanced Research Projects Agency. Revolutionary Enhancement of Visibility by Exploiting Active Light-fields Program (Contract HR0011- 16-C-0030)

This work presents a novel interactive system for simple garment composition and surface patterning. Our approach makes it easier for casual users to customize machine-knitted garments, while enabling more advanced users to design their own composable templates. Our tool combines ideas from CAD software and image editing: it allows the composition of (1) parametric knitted primitives, and (2) stitch pattern layers with different resampling behaviours. By leveraging the regularity of our primitives, our tool enables interactive customization with automated layout and real-time patterning feedback. We show a variety of garments and patterns created with our tool, and highlight our ability to transfer shape and pattern customizations between users., Comment: Project page: http://knitskel.csail.mit.edu

Many applications perform frequent scatter update operations to large data structures. For example, in push-style graph algorithms, processing each vertex requires updating the data of all its neighbors. Neighbors are often scattered over the whole graph, so these scatter updates have poor spatial and temporal locality. In current systems, scatter updates suffer high synchronization costs and high memory traffic. These drawbacks make push-style execution unattractive, and, when algorithms allow it, programmers gravitate towards pull-style implementations based on gather reads instead. We present PHI, a push cache hierarchy that makes scatter updates synchronization- and bandwidth-efficient. PHI adds support for pushing sparse, commutative updates from cores towards main memory. PHI adds simple compute logic at each cache level to buffer and coalesce these commutative updates throughout the hierarchy. This avoids synchronization, exploits temporal locality, and produces a load-balanced execution. Moreover, PHI exploits spatial locality by selectively deferring updates with poor spatial locality, batching them to achieve sequential main memory transfers. PHI is the first system to leverage both the temporal and spatial locality benefits of commutative scatter updates, some of which do not apply to gather reads. As a result, PHI not only makes push algorithms efficient, but makes them consistently faster than pull ones. We evaluate PHI on graph algorithms and other sparse applications processing large inputs. PHI improves performance by 4.7× on average (and by up to 11×), and reduces memory traffic by 2× (and by up to 5×).

Probabilistic atlas priors have been commonly used to derive adaptive and robust brain MRI segmentation algorithms. Widely-used neuroimage analysis pipelines rely heavily on these techniques, which are often computationally expensive. In contrast, there has been a recent surge of approaches that leverage deep learning segmentation tools that are computationally efficient at test time. However, most of these strategies rely on supervised learning from manually annotated images and are therefore sensitive to the intensity profiles in the training dataset. A deep learning-based segmentation model for a new image dataset (e.g., of different contrast), usually requires a new labeled training dataset, which can be prohibitively expensive, or suboptimal ad hoc adaptation or augmentation approaches. In this paper, we propose an alternative strategy that combines conventional probabilistic atlas-based segmentation with deep learning, enabling training of a segmentation model for new MRI scans without the need for any manually segmented images. Our experiments include thousands of brain MRI scans and demonstrate that the proposed method achieves good accuracy for a brain MRI segmentation task for different MRI contrasts, requiring only approximately 15 s at test time on a GPU., European Research Council (Starting Grant 677697), BRAIN Initiative. Cell Census Network (Grant U01-MH117023), National Institutes of Health (U.S.) (Grants R21-AG050122, P41-EB015902, 5U01-MH093765, R01LM012719, R01AG053949, and 1R21AG050122), National Institute of Biomedical Imaging and Bioengineering (U.S.) (Grants P41-EB015896, 1R01-EB023281, R01-EB006758, R21-EB018907, R01-EB019956), National Institute on Aging (Grants 5R01-AG008122, R01-AG016495), National Institute of Diabetes and Digestive and Kidney Diseases (U.S.) (Grant 1R21-DK-108277-01), National Institute of Neurological Diseases and Stroke (Grants R01-NS0525851, R21-NS072652, R01-NS070963, R01-NS083534, 5U01-NS086625, 5U24-NS10059103), National Science Foundation (U.S.). NeuroNex Grant (Grant 1707312), National Science Foundation (U.S.). Career (1748377), National Institutes of Health (U.S.). Shared Instrumentation Grant Program (Grants 1S10RR023401, 1S10RR019307, and 1S10RR023043)

Generative Adversarial Networks (GANs) typically learn a distribution of images in a large image dataset, and are then able to generate new images from this distribution. However, each natural image has its own internal statistics, captured by its unique distribution of patches. In this paper we propose an ``Internal GAN'' (InGAN) -- an image-specific GAN -- which trains on a single input image and learns its internal distribution of patches. It is then able to synthesize a plethora of new natural images of significantly different sizes, shapes and aspect-ratios – all with the same internal patch-distribution (same ``DNA'') as the input image. In particular, despite large changes in global size/shape of the image, all elements inside the image maintain their local size/shape. InGAN is fully unsupervised, requiring no additional data other than the input image itself. Once trained on the input image, it can remap the input to any size or shape in a single feedforward pass, while preserving the same internal patch distribution. InGAN provides a unified framework for a variety of tasks, bridging the gap between textures and natural images.

This paper presents RF-Avatar, a neural network model that can estimate 3D meshes of the human body in the presence of occlusions, baggy clothes, and bad lighting conditions. We leverage that radio frequency (RF) signals in the WiFi range traverse clothes and occlusions and bounce off the human body. Our model parses such radio signals and recovers 3D body meshes. Our meshes are dynamic and smoothly track the movements of the corresponding people. Further, our model works both in single and multi-person scenarios. Inferring body meshes from radio signals is a highly under-constrained problem. Our model deals with this challenge using: 1) a combination of strong and weak supervision, 2) a multi-headed self-attention mechanism that attends differently to temporal information in the radio signal, and 3) an adversarially trained temporal discriminator that imposes a prior on the dynamics of human motion. Our results show that RF-Avatar accurately recovers dynamic 3D meshes in the presence of occlusions, baggy clothes, bad lighting conditions, and even through walls.

This paper introduces Perennial, a framework for verifying concurrent, crash-safe systems. Perennial extends the Iris concurrency framework with three techniques to enable crash-safety reasoning: recovery leases, recovery helping, and versioned memory. To ease development and deployment of applications, Perennial provides Goose, a subset of Go and a translator from that subset to a model in Perennial with support for reasoning about Go threads, data structures, and file-system primitives. We implemented and verified a crash-safe, concurrent mail server using Perennial and Goose that achieves speedup on multiple cores. Both Perennial and Iris use the Coq proof assistant, and the mail server and the framework’s proofs are machine checked., NSF (Awards CNS-1563763 and CCF- 1836712)

Part of the Lecture Notes in Computer Science book series (LNCS, volume 11767), The performance and diagnostic utility of magnetic resonance imaging (MRI) in pregnancy is fundamentally constrained by fetal motion. Motion of the fetus, which is unpredictable and rapid on the scale of conventional imaging times, limits the set of viable acquisition techniques to single-shot imaging with severe compromises in signal-to-noise ratio and diagnostic contrast, and frequently results in unacceptable image quality. Surprisingly little is known about the characteristics of fetal motion during MRI and here we propose and demonstrate methods that exploit a growing repository of MRI observations of the gravid abdomen that are acquired at low spatial resolution but relatively high temporal resolution and over long durations (10–30 min). We estimate fetal pose per frame in MRI volumes of the pregnant abdomen via deep learning algorithms that detect key fetal landmarks. Evaluation of the proposed method shows that our framework achieves quantitatively an average error of 4.47 mm and 96.4% accuracy (with error less than 10 mm). Fetal pose estimation in MRI time series yields novel means of quantifying fetal movements in health and disease, and enables the learning of kinematic models that may enhance prospective mitigation of fetal motion artifacts during MRI acquisition., NIH (Grants U01HD087211, R01EB01733, NIBIB NAC P41EB015902 and NICHD U01HD087211)

Recent attacks have broken process isolation by exploiting microarchitectural side channels that allow indirect access to shared microarchitectural state. Enclaves strengthen the process abstraction to restore isolation guarantees. We propose MI6, an aggressively speculative out-of-order processor capable of providing secure enclaves under a threat model that includes an untrusted OS and an attacker capable of mounting any software attack currently considered practical, including those utilizing control flow mis-speculation. MI6 is inspired by Sanctum [16] and extends its isolation guarantee to more realistic memory hierarchy. It also introduces a purge instruction, which is used only when a secure process is (de)scheduled, and implements it for a complex processor microarchitecture. We model the performance impact of enclaves in MI6 through FPGA emulation on AWS F1 FPGAs by running SPEC CINT2006 benchmarks as enclaves within an untrusted Linux OS. Security comes at the cost of approximately 16.4% average slowdown for protected programs., National Science Foundation (U.S.) (Grant CNS-1413920), United States. Space and Naval Warfare Systems Command (Contract N66001-15-C-4066), United States. Defense Advanced Research Projects Agency (Grant HR001118C0018)

Summary: To reveal post-traumatic stress disorder (PTSD) genetic risk influences on tissue-specific gene expression, we use brain and non-brain transcriptomic imputation. We impute genetically regulated gene expression (GReX) in 29,539 PTSD cases and 166,145 controls from 70 ancestry-specific cohorts and identify 18 significant GReX-PTSD associations corresponding to specific tissue-gene pairs. The results suggest substantial genetic heterogeneity based on ancestry, cohort type (military versus civilian), and sex. Two study-wide significant PTSD associations are identified in European and military European cohorts; ZNF140 is predicted to be upregulated in whole blood, and SNRNP35 is predicted to be downregulated in dorsolateral prefrontal cortex, respectively. In peripheral leukocytes from 175 marines, the observed PTSD differential gene expression correlates with the predicted differences for these individuals, and deployment stress produces glucocorticoid-regulated expression changes that include downregulation of both ZNF140 and SNRNP35. SNRNP35 knockdown in cells validates its functional role in U12-intron splicing. Finally, exogenous glucocorticoids in mice downregulate prefrontal Snrnp35 expression.

Understanding people's actions and interactions typically depends on seeing them. Automating the process of action recognition from visual data has been the topic of much research in the computer vision community. But what if it is too dark, or if the person is occluded or behind a wall? In this paper, we introduce a neural network model that can detect human actions through walls and occlusions, and in poor lighting conditions. Our model takes radio frequency (RF) signals as input, generates 3D human skeletons as an intermediate representation, and recognizes actions and interactions of multiple people over time. By translating the input to an intermediate skeleton-based representation, our model can learn from both vision-based and RF-based datasets, and allow the two tasks to help each other. We show that our model achieves comparable accuracy to vision-based action recognition systems in visible scenarios, yet continues to work accurately when people are not visible, hence addressing scenarios that are beyond the limit of today's vision-based action recognition., ICCV 2019. The first two authors contributed equally to this paper

Part of the Lecture Notes in Computer Science book series (LNCS, volume 11767)., We present a volumetric mesh-based algorithm for flattening the placenta to a canonical template to enable effective visualization of local anatomy and function. Monitoring placental function in vivo promises to support pregnancy assessment and to improve care outcomes. We aim to alleviate visualization and interpretation challenges presented by the shape of the placenta when it is attached to the curved uterine wall. To do so, we flatten the volumetric mesh that captures placental shape to resemble the well-studied ex vivo shape. We formulate our method as a map from the in vivo shape to a flattened template that minimizes the symmetric Dirichlet energy to control distortion throughout the volume. Local injectivity is enforced via constrained line search during gradient descent. We evaluate the proposed method on 28 placenta shapes extracted from MRI images in a clinical study of placental function. We achieve sub-voxel accuracy in mapping the boundary of the placenta to the template while successfully controlling distortion throughout the volume. We illustrate how the resulting mapping of the placenta enhances visualization of placental anatomy and function. Our implementation is freely available at https://github.com/mabulnaga/placenta-flattening., NIH/NIBIB/NAC (Grant P41EB015902), NIH/NICHD (Grant U01HD087211), NSF (Grant IIS-1838071), Air Force Office of Scientific Research (Award FA9550-19-1-0319)

The TADPOLE Challenge compares the performance of algorithms at predicting the future evolution of individuals at risk of Alzheimer's disease. TADPOLE Challenge participants train their models and algorithms on historical data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) study. Participants are then required to make forecasts of three key outcomes for ADNI-3 rollover participants: clinical diagnosis, ADAS-Cog 13, and total volume of the ventricles -- which are then compared with future measurements. Strong points of the challenge are that the test data did not exist at the time of forecasting (it was acquired afterwards), and that it focuses on the challenging problem of cohort selection for clinical trials by identifying fast progressors. The submission phase of TADPOLE was open until 15 November 2017; since then data has been acquired until April 2019 from 219 subjects with 223 clinical visits and 150 Magnetic Resonance Imaging (MRI) scans, which was used for the evaluation of the participants' predictions. Thirty-three teams participated with a total of 92 submissions. No single submission was best at predicting all three outcomes. For diagnosis prediction, the best forecast (team Frog), which was based on gradient boosting, obtained a multiclass area under the receiver-operating curve (MAUC) of 0.931, while for ventricle prediction the best forecast (team EMC1), which was based on disease progression modelling and spline regression, obtained mean absolute error of 0.41% of total intracranial volume (ICV). For ADAS-Cog 13, no forecast was considerably better than the benchmark mixed effects model (BenchmarkME), provided to participants before the submission deadline. Further analysis can help understand which input features and algorithms are most suitable for Alzheimer's disease prediction and for aiding patient stratification in clinical trials., 10 pages, 1 figure, 4 tables. arXiv admin note: substantial text overlap with arXiv:1805.03909

Part of the Lecture Notes in Computer Science book series (LNCS, volume 11846), We present BrainPainter, a software that automatically generates images of highlighted brain structures given a list of numbers corresponding to the output colours of each region. Compared to existing visualisation software (i.e. Freesurfer, SPM, 3D Slicer), BrainPainter has three key advantages: (1) it does not require the input data to be in a specialised format, allowing BrainPainter to be used in combination with any neuroimaging analysis tools, (2) it can visualise both cortical and subcortical structures and (3) it can be used to generate movies showing dynamic processes, e.g. propagation of pathology on the brain. We highlight three use cases where BrainPainter was used in existing neuroimaging studies: (1) visualisation of the degree of atrophy through interpolation along a user-defined gradient of colours, (2) visualisation of the progression of pathology in Alzheimer’s disease as well as (3) visualisation of pathology in subcortical regions in Huntington’s disease. Moreover, through the design of BrainPainter we demonstrate the possibility of using a powerful 3D computer graphics engine such as Blender to generate brain visualisations for the neuroscience community. Blender’s capabilities, e.g. particle simulations, motion graphics, UV unwrapping, raster graphics editing, raytracing and illumination effects, open a wealth of possibilities for brain visualisation not available in current neuroimaging software. BrainPainter (Source code: https://github.com/mrazvan22/brain-coloring ) is customisable, easy to use, and can run straight from the web browser: http://brainpainter.csail.mit.edu. It can be used to visualise biomarker data from any brain imaging modality, or simply to highlight a particular brain structure for e.g. anatomy courses., NIH (Grants NIBIB NAC P41EB015902 and NINDS R01NS086905)