Fundamental Tradeoff Research Articles

Repression Works (Just Not in Moderation)

Why does government violence deter political challengers in one context but inflame them in the next? This paper argues that repression increases opposition activity at low and moderate levels but decreases it in the extreme. There is a threshold level of violence, where the opposition becomes unable to recruit new members, and the rebellion unravels—even if the government kills more innocents. We find empirical support for this proposition in disaggregated data from Chechnya and a meta-analysis of sub-national conflict dynamics in 71 countries. The data suggest that a threshold exists, but the level of violence needed to reach it varies. Many governments, thankfully, are unable or unwilling to go that far. We explore conditions under which this threshold may be higher or lower and highlight a fundamental trade-off between reducing government violence and preserving civil liberties.JEL ClassificationD74, F51, H56.

Influence of processing on the microstructural evolution and multiscale hardness in titanium carbonitrides (TiCN) produced via field assisted sintering technology

Titanium carbonitride (TiCN) is an advanced, high-performance hard ceramic of great commercial importance that has been widely developed and employed. Nonetheless, it has only been in recent years that binderless titanium carbonitride bulk ceramics have been successfully fabricated using field-assisted sintering technology (FAST). However, the underlying structure-processing-property-performance relationships have yet to be fully evaluated, especially concerning indentation hardness of these materials across a broad range of loads and deformation length scales. In this work we aim to address these fundamental relationships and characterize the multiscale hardness phenomena in detail. It was found that the effects of soak temperature and time directly impacted the sintered microstructure and were reflected in the observed mechanical properties over various loads. Valuable insight into the load-dependence of hardness distributions, sensitivity/correlation with elastoplastic parameters, and multiscale parameterization were developed using micro-/nanoindentation. Particularly, the load-dependent hardness sensitivity and resolvability demonstrate a fundamental tradeoff with respect to the manifested mechanical response influenced by the presence of underlying heterogeneities. These new insights relating the interplay of compositional/microstructural evolution with FAST processing parameters and multiscale hardness are an important step in advancing next-generation hard ceramics.

Opportunities and Limitations in Broadband Sensing

Detecting a signal at an unknown frequency is a common task, arising in settings from dark-matter detection to magnetometry. For any detection protocol, the precision achieved depends on the frequency of the signal and can be quantified by the quantum Fisher information (QFI). To study limitations in broadband sensing, we introduce the integrated quantum Fisher information and derive inequality bounds that embody fundamental trade-offs in any sensing protocol. Our inequalities show that sensitivity in one frequency range must come at the cost of reduced sensitivity elsewhere. For many protocols, including those with small phase accumulation and those consisting of $\ensuremath{\pi}$ pulses, we find that the integrated quantum Fisher information scales linearly with $T$. We also find protocols with substantial phase accumulation that can have integrated QFI that grows quadratically with $T$ and prove that this scaling is asymptotically optimal. These protocols may allow the very rapid detection of a signal with unknown frequency over a very wide bandwidth. We discuss the implications of these results for a wide variety of contexts, including dark-matter searches and dynamical decoupling. Thus we establish fundamental limitations on the broadband detection of signals and highlight their consequences.

On the Value of Dynamism in Transit Networks

The rise of on-demand mobility technologies over the past decade has sparked interest in the integration of traditional transit and on-demand systems. One of the main reasons behind this is the potential to address a fundamental trade-off in transit: the ridership versus coverage dilemma. However, unlike purely fixed systems or purely on-demand systems, integrated systems are not well understood; their planning and operational problems are significantly more challenging, and their broader implications are the source of a heated debate. Motivated by this debate, we introduce the dynamicity gap, a general concept that quantifies the attainable benefit of allowing (but not requiring) dynamic components in the response strategy to a multistage optimization problem. Although computing the dynamicity gap exactly may be intractable, we develop an analytical framework with which to approximate it as a function of problem input parameters. The framework allows us to certify the value of dynamism (i.e., a dynamicity gap greater than one) for certain combinations of problem input parameters. We showcase our approach with two sets of computational experiments, from which we gain both qualitative and quantitative insights about the settings in which the integration of transit and on-demand systems may certifiably be a worthwhile investment.Funding: This work was partially supported by the National Science Foundation [Grants DMS-1839346 and CNS-1952011]. Part of this research was performed while the authors were visiting the Institute for Pure and Applied Mathematics, which is supported by the National Science Foundation [Grant DMS-1925919].Supplemental Material: The online appendix is available at https://doi.org/10.1287/trsc.2022.1193 .

Fundamental trade-off between the speed of light and the Fano factor of photon current in three-level lambda systems

Electromagnetically induced slow-light medium is a promising system for quantum memory devices, but controlling its noise level remains a major challenge to overcome. This work considers the simplest model for such medium, comprised of three-level Λ-systems interacting with bosonic bath, and provides a new fundamental trade-off relation in light–matter interaction between the group velocity of light and the Fano factor of photon current due to radiative transitions. Considering the steady state limits of a newly derived Lindblad-type equation, we find that the Fano factor of the photon current maximizes to 3 at the minimal group velocity of light, which holds true universally regardless of detailed values of parameters characterizing the medium.

Constrained regret minimization for multi-criterion multi-armed bandits

We consider a stochastic multi-armed bandit setting and study the problem of constrained regret minimization over a given time horizon. Each arm is associated with an unknown, possibly multi-dimensional distribution, and the merit of an arm is determined by several, possibly conflicting attributes. The aim is to optimize a ‘primary’ attribute subject to user-provided constraints on other ‘secondary’ attributes. We assume that the attributes can be estimated using samples from the arms’ distributions, and that the estimators enjoy suitable concentration properties. We propose an algorithm called Con-LCB that guarantees a logarithmic regret, i.e., the average number of plays of all non-optimal arms is at most logarithmic in the horizon. The algorithm also outputs a boolean flag that correctly identifies, with high probability, whether the given instance is feasible/infeasible with respect to the constraints. We also show that Con-LCB is optimal within a universal constant, i.e., that more sophisticated algorithms cannot do much better universally. Finally, we establish a fundamental trade-off between regret minimization and feasibility identification. Our framework finds natural applications, for instance, in financial portfolio optimization, where risk constrained maximization of expected return is meaningful.

Odyssey: A Journey in the Land of Distributed Data Series Similarity Search

This paper presents Odyssey, a novel distributed data-series processing framework that efficiently addresses the critical challenges of exhibiting good speedup and ensuring high scalability in data series processing by taking advantage of the full computational capacity of modern distributed systems comprised of multi-core servers. Odyssey addresses a number of challenges in designing efficient and highly-scalable distributed data series index, including efficient scheduling, and load-balancing without paying the prohibitive cost of moving data around. It also supports a flexible partial replication scheme, which enables Odyssey to navigate through a fundamental trade-off between data scalability and good performance during query answering. Through a wide range of configurations and using several real and synthetic datasets, our experimental analysis demonstrates that Odyssey achieves its challenging goals.

Frequency domain LPV controller synthesis for a positioning system with uncertain scheduling parameters

Ever-increasing demands from the industry require better performance out of the same system to achieve a competitive advantage. In systems functioning at different operating points, a fundamental trade-off between robustness and performance can limit the achieved performance. Linear Parameter Varying (LPV) controller synthesis can overcome this issue by adapting the controller parameters with the operating points. This requires a parametric LPV model of the system, which in many cases is difficult or costly to obtain. Recently, LPV controller design methods based on the frequency-domain data in different operating points have shown to result in great performance without the need of a parametric LPV model. In these approaches, however, the scheduling parameters are assumed to be measured exactly. In this paper, a new approach is proposed to design LPV controllers with slowly varying uncertain scheduling parameters, using only the frequency response data of a SISO system around different operating points. The proposed approach is applied on a rotary table, a machine used in the manufacturing industry. The dynamics of the rotary table depend on the inertia of the object mounted on-top which cannot be directly measured. Since the scheduling parameter cannot be measured, it is estimated using a short sequence of data that leads to an inexact estimation with interval uncertainty. For the system considered, the LPV controller designed by the proposed approach shows clear improvement over the state-of-the-art robust controller synthesis approaches.

Space-Time Super-Resolution for Light Field Videos.

Light field (LF) cameras suffer from a fundamental trade-off between spatial and angular resolutions. Additionally, due to the significant amount of data that needs to be recorded, the Lytro ILLUM, a modern LF camera, can only capture three frames per second. In this paper, we consider space-time super-resolution (SR) for LF videos, aiming at generating high-resolution and high-frame-rate LF videos from low-resolution and low-frame-rate observations. Extending existing space-time video SR methods to this task directly will meet two key challenges: 1) how to re-organize sub-aperture images (SAIs) efficiently and effectively given highly redundant LF videos, and 2) how to aggregate complementary information between multiple SAIs and frames considering the coherence in LF videos. To address the above challenges, we propose a novel framework for space-time super-resolving LF videos for the first time. First, we propose a novel Multi-Scale Dilated SAI Re-organization strategy for re-organizing SAIs into auxiliary view stacks with decreasing resolution as the Chebyshev distance in the angular dimension increases. In particular, the auxiliary view stack with original resolution preserves essential visual details, while the down-scaled view stacks capture long-range contextual information. Second, we propose the Multi-Scale Aggregated Feature extractor and the Angular-Assisted Feature Interpolation module to utilize and aggregate information from the spatial, angular, and temporal dimensions in LF videos. The former aggregates similar contents from different SAIs and frames for subsequent reconstruction in a disparity-free manner at the feature level, whereas the latter interpolates intermediate frames temporally by implicitly aggregating geometric information. Compared to other potential approaches, experimental results demonstrate that the reconstructed LF videos generated by our framework achieve higher reconstruction quality and better preserve the LF parallax structure and temporal consistency. The implementation code is available at https://github.com/zeyuxiao1997/LFSTVSR.

The Generic Drug Trilemma

More than 90% of prescriptions dispensed in the United States each year are for off-patent drugs. Yet the bulk of scholarship on prescription drug policy focuses on patented drugs. Discussions of prescription drugs are typically oriented around the “innovation-access dilemma”—the trade-off between stronger patent-based incentives for innovators and higher prices for purchasers of patented products. But for drugs in the “patent afterlife”—the period after patent protection and other forms of market exclusivity have expired—the innovation-access dilemma is not the fundamental policy trade-off. Higher prices do not necessarily redound to the benefit of innovators, and price is not the only significant impediment to access: drug shortages—often triggered by safety concerns—also prevent patients from obtaining the medicines they need.This chapter seeks to provide scholars and policy makers with a unifying framework for analyzing the variegated challenges of the pharmaceutical patent afterlife. We argue that the key trade-offs in off-patent drug policy take the form of a trilemma—the three corners of which are price, quantity, and quality (i.e., safety and efficacy). The generally accepted goal of off-patent drug policy is to facilitate (1) low prices and (2) sufficient quantities of drugs that are (3) equivalent or similar to brand-name drugs approved by the Food and Drug Administration. But policies that improve outcomes along one or two of those dimensions typically entail some sacrifice along the third.The trilemma framing yields several analytical payoffs. First, it sheds light on the root causes of puzzling problems plaguing some off-patent drug markets, such as sudden price spikes and persistent mismatches between supply and demand. Second, it draws attention to the noninnovation costs of recent drug pricing reform proposals—costs that are often overlooked amid the emphasis on innovation versus access. Finally, it motivates a search for solutions that can transcend the trilemma and optimize along all three dimensions of off-patent drug policy. That search brings us, full circle, back to innovation and access—the concerns that dominate discussions of pharmaceutical policy at the beginning of the patent life. But this time, the focus is on innovation—and, specifically, manufacturing innovation—in service of rather than in tension with access.

When You Can't Afford to Wait for a Job: The Role of Time Discounting for Own-Account Workers in Developing Countries

Frictional labor markets impose a fundamental trade-off: individuals may work on their own at any time, but can only take a potentially better-paid wage job after spending some time looking for it, suggesting that intertemporal considerations affect how people choose their occupation. We formalize this intuition under the job search framework and show that a sufficiently high subjective discount rate can justify the choice for own-account work even when it pays less than wage work. With this simple model, we estimate a lower bound for the discount rate that is implicit in the occupational choice of urban own-accountworkers in Brazil. We find that at least 65% of those workers appear to discount the future at rates superior to those available in the credit market, which suggests constrained occupational choice. Finally, we show that the es- timated time preference lower bound is positively associated with food, clothing, and housing deprivation.

RDMA Congestion Control: It Is Only for the Compliant

Remote direct memory access (RDMA) networks enable low latency and low central processing unit utilization, and their widespread adoption in datacenters enables improved application performance. However, there are performance isolation concerns for RDMA deployed in a shared cloud environment. In particular, congestion control enforcement and congestion control algorithms in RDMA make the network susceptible to performance hacking attacks, which give the attacker extra bandwidth and cause severe congestion in the network. These attacks can increase short flow completion times by several orders of magnitude. We surface a fundamental tradeoff in congestion control between short flow completion time and performance isolation. We discuss this tradeoff and how existing approaches do not provide a robust solution. We also advocate that researchers incorporate performance isolation concerns into the design and evaluation of congestion control.

DIANA: An End-to-End Hybrid DIgital and ANAlog Neural Network SoC for the Edge

DIgital-ANAlog (DIANA), a heterogeneous multi-core accelerator, combines a reduced instruction set computer - five (RISC-V) host processor with an analog in-memory computing (AIMC) artificial intelligence (AI) accelerator and a digital reconfigurable deep neural network (DNN) accelerator in a single system-on-chip (SoC) to support a wide variety of neural network (NN) workloads. AIMC cores can bring extreme computational parallelism and efficiency at the expense of accuracy and dataflow flexibility. Digital AI co-processors, on the other hand, guarantee accuracy through deterministic compute, but cannot achieve the same computational density and efficiency. DIANA exploits this fundamental tradeoff by integrating both types of cores in a shared and optimized memory system, to enable seamless execution of the workloads on the parallel cores. The system’s performance benefits further from pipelined parallel execution across both accelerator cores and enhanced AIMC spatial unrolling techniques, leading to drastically reduced execution latency and reduced memory footprints. The design has been implemented in a 22-nm technology and achieves peak efficiencies of 600 TOP/s/W for the AIMC core (I/W/O: 7/1.5/6 bit) and 14 TOP/s/W (I/W/O: 8/8/8 bit) for the digital accelerator, respectively. End-to-end performance evaluation of CIFAR-10 and ImageNet classification workloads is carried out on the chip, reporting 7.02 and 5.56 TOP/s/W, respectively, at the system level.

The Right to Water and Sustainable Consumption in EU Law

The right to water has gathered momentum in recent years in Europe, having become the subject of the first European citizens’ initiative and emerging as a human right. The right to water, however, is in constant, and not always linear, evolution as it faces fundamental trade-offs; on the one hand, access to clean and affordable water is essential to ensure a basic standard of living. On the other, water is an increasingly limited resource, so unfettered access to it increases the threat of scarcity and pollution. This article examines the interplay between the right to water and sustainable consumption objectives, exploring how innovation in regulation and best practice could reduce the risks to health and water scarcity. As water regulation affects multiple areas of law, the article will examine the right to water and sustainable consumption from a human rights’ angle and taking a consumer law and environmental protection perspective. A particular focus will be on the Water Framework Directive and the recently revised Drinking Water Directive, examining incentives that promote water rights and sustainable water use. Key regulatory instruments will be evaluated, ranging from information and education tools to economic and social incentives. Finally, the article will propose new measures to align the right to water with the objective of sustainable consumption.

Top-Illuminated Avalanche Photodiodes With Cascaded Multiplication Layers for High-Speed and Wide Dynamic Range Performance

In this work, a novel top-illuminated avalanche photodiode (APD) with a In <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0.52</sub> Al <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0.48</sub> As multiplication (M-) layer is demonstrated. The cascaded M-layer design combined with the unique p-side up mesa structure allows relaxation of the fundamental trade-off between the gain-bandwidth product and the dark current. This leads to the simultaneous high-responsivity, high-speed, high-saturation-power, and low-dark current characteristics of our APDs. At around 0.9 V <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">br</sub> operation, the demonstrated device with its simple top-illuminated structure and large active window (mesa) diameter of 14 (24) μm exhibits a high responsivity (2.23 A/W), wide optical-to-electrical bandwidth (30 GHz), large gain-bandwidth product (270 GHz), low dark current (∼200 nA), and a saturation current as high as 11 mA. The excellent performance of this APD structure opens up new possibilities to further enhance the sensitivity performance of receivers for coherent communications or 106 Gbit/sec PAM-4 applications.

Fundamental Limitation on the Detectability of Entanglement.

Entanglement detection is essential in quantum information science and quantum many-body physics. It has been proved that entanglement exists almost surely for a random quantum state, while the realizations of effective entanglement criteria usually consume exponentially many resources with regard to system size or qubit number, and efficient criteria often perform poorly without prior knowledge. This fact implies a fundamental limitation might exist in the detectability of entanglement. In this work, we formalize this limitation as a fundamental trade-off between the efficiency and effectiveness of entanglement criteria via a systematic method to evaluate the detection capability of entanglement criteria theoretically. For a system coupled to an environment, we prove that any entanglement criterion needs exponentially many observables to detect the entanglement effectively when restricted to single-copy operations. Otherwise, the detection capability of the criterion will decay double exponentially. Furthermore, if multicopy joint measurements are allowed, the effectiveness of entanglement detection can be exponentially improved, which implies a quantum advantage in entanglement detection problems. Our results may shed light on why quantum phenomena are difficult to observe in large noisy systems.

Bio-mimetic high-speed target localization with fused frame and event vision for edge application.

Evolution has honed predatory skills in the natural world where localizing and intercepting fast-moving prey is required. The current generation of robotic systems mimics these biological systems using deep learning. High-speed processing of the camera frames using convolutional neural networks (CNN) (frame pipeline) on such constrained aerial edge-robots gets resource-limited. Adding more compute resources also eventually limits the throughput at the frame rate of the camera as frame-only traditional systems fail to capture the detailed temporal dynamics of the environment. Bio-inspired event cameras and spiking neural networks (SNN) provide an asynchronous sensor-processor pair (event pipeline) capturing the continuous temporal details of the scene for high-speed but lag in terms of accuracy. In this work, we propose a target localization system combining event-camera and SNN-based high-speed target estimation and frame-based camera and CNN-driven reliable object detection by fusing complementary spatio-temporal prowess of event and frame pipelines. One of our main contributions involves the design of an SNN filter that borrows from the neural mechanism for ego-motion cancelation in houseflies. It fuses the vestibular sensors with the vision to cancel the activity corresponding to the predator's self-motion. We also integrate the neuro-inspired multi-pipeline processing with task-optimized multi-neuronal pathway structure in primates and insects. The system is validated to outperform CNN-only processing using prey-predator drone simulations in realistic 3D virtual environments. The system is then demonstrated in a real-world multi-drone set-up with emulated event data. Subsequently, we use recorded actual sensory data from multi-camera and inertial measurement unit (IMU) assembly to show desired working while tolerating the realistic noise in vision and IMU sensors. We analyze the design space to identify optimal parameters for spiking neurons, CNN models, and for checking their effect on the performance metrics of the fused system. Finally, we map the throughput controlling SNN and fusion network on edge-compatible Zynq-7000 FPGA to show a potential 264 outputs per second even at constrained resource availability. This work may open new research directions by coupling multiple sensing and processing modalities inspired by discoveries in neuroscience to break fundamental trade-offs in frame-based computer vision.

Structured cerebellar connectivity supports resilient pattern separation.

The cerebellum is thought to helpdetect and correct errors between intended and executed commands1,2 and is critical for social behaviours, cognition and emotion3-6. Computations for motor control must be performed quickly to correct errors in real time and should be sensitive to small differences between patterns for fine error correction while being resilient to noise7. Influential theories of cerebellar information processing have largely assumed random network connectivity, which increases the encoding capacity of the network's first layer8-13. However, maximizing encoding capacity reduces the resilience to noise7. To understand how neuronal circuits address this fundamental trade-off, we mapped the feedforward connectivity in the mouse cerebellar cortex using automated large-scale transmission electron microscopy and convolutional neural network-based image segmentation. We found that both the input and output layers of the circuit exhibit redundant and selective connectivity motifs, which contrast with prevailing models. Numerical simulations suggest that these redundant, non-random connectivity motifs increase the resilience to noise at a negligible cost to the overall encoding capacity. This work reveals how neuronal network structure can support a trade-off between encoding capacity and redundancy, unveiling principles of biological network architecture with implications for the design of artificial neural networks.

Intergenerational consequences of gradual pension reforms

Balancing the government budget in an aging economy may require adjusting gradually pension benefits. Such policy change can take two forms: adjusting the accrual rate (the rate at which individuals built-up pension entitlements while working) or the indexation rate (the rate a which accrued entitlements are linked to nominal wage growth). We compare the consequences of such gradual policies across cohorts. We identify a fundamental generational trade-off between democracy and equality. In particular, we show that for Belgium, 80% of the population alive at the time of the reform prefers the accrual to the indexation reform, with the implication that the youngest half of the population would bear 85% of the total adjustment cost. The indexation reform provides more generational equality because the phasing in over time has larger base and thus benefit cut can be smaller per capita. We then consider other reforms improving the generational equality, showing that all those reforms fail to gain majority support. Finally, considering labor incentives, we show that the indexation reform is also more efficient than the accrual reform. Efficiency meets generational equality.

Attractor and integrator networks in the brain.

In this Review, we describe the singular success of attractor neural network models in describing how the brain maintains persistent activity states for working memory, corrects errors and integrates noisy cues. We consider the mechanisms by which simple and forgetful units can organize to collectively generate dynamics on the long timescales required for such computations. We discuss the myriad potential uses of attractor dynamics for computation in the brain, and showcase notable examples of brain systems in which inherently low-dimensional continuous-attractor dynamics have been concretely and rigorously identified. Thus, it is now possible to conclusively state that the brain constructs and uses such systems for computation. Finally, we highlight recent theoretical advances in understanding how the fundamental trade-offs between robustness and capacity and between structure and flexibility can be overcome by reusing and recombining the same set of modular attractors for multiple functions, so they together produce representations that are structurally constrained and robust but exhibit high capacity and are flexible.