-
"Zero Cost'' Majority Attacks on Permissionless Blockchains
Authors:
Joshua S. Gans,
Hanna Halaburda
Abstract:
The core premise of permissionless blockchains is their reliable and secure operation without the need to trust any individual agent. At the heart of blockchain consensus mechanisms is an explicit cost (whether work or stake) for participation in the network and the opportunity to add blocks to the blockchain. A key rationale for that cost is to make attacks on the network, which could be theoreti…
▽ More
The core premise of permissionless blockchains is their reliable and secure operation without the need to trust any individual agent. At the heart of blockchain consensus mechanisms is an explicit cost (whether work or stake) for participation in the network and the opportunity to add blocks to the blockchain. A key rationale for that cost is to make attacks on the network, which could be theoretically carried out if a majority of nodes were controlled by a single entity, too expensive to be worthwhile. We demonstrate that a majority attacker can successfully attack with a {\em negative cost}, which shows that the protocol mechanisms are insufficient to create a secure network, and emphasizes the importance of socially driven mechanisms external to the protocol. At the same time, negative cost enables a new type of majority attack that is more likely to elude external scrutiny.
△ Less
Submitted 12 August, 2023;
originally announced August 2023.
-
The Economics of Augmented and Virtual Reality
Authors:
Joshua Gans,
Abhishek Nagaraj
Abstract:
This paper explores the economics of Augmented Reality (AR) and Virtual Reality (VR) technologies within decision-making contexts. Two metrics are proposed: Context Entropy, the informational complexity of an environment, and Context Immersivity, the value from full immersion. The analysis suggests that AR technologies assist in understanding complex contexts, while VR technologies provide access…
▽ More
This paper explores the economics of Augmented Reality (AR) and Virtual Reality (VR) technologies within decision-making contexts. Two metrics are proposed: Context Entropy, the informational complexity of an environment, and Context Immersivity, the value from full immersion. The analysis suggests that AR technologies assist in understanding complex contexts, while VR technologies provide access to distant, risky, or expensive environments. The paper provides a framework for assessing the value of AR and VR applications in various business sectors by evaluating the pre-existing context entropy and context immersivity. The goal is to identify areas where immersive technologies can significantly impact and distinguish those that may be overhyped.
△ Less
Submitted 26 May, 2023;
originally announced May 2023.
-
Mechanism Design Approaches to Blockchain Consensus
Authors:
Joshua S. Gans,
Richard Holden
Abstract:
Blockchain consensus is a state whereby each node in a network agrees on the current state of the blockchain. Existing protocols achieve consensus via a contest or voting procedure to select one node as a dictator to propose new blocks. However, this procedure can still lead to potential attacks that make consensus harder to achieve or lead to coordination issues if multiple, competing chains (i.e…
▽ More
Blockchain consensus is a state whereby each node in a network agrees on the current state of the blockchain. Existing protocols achieve consensus via a contest or voting procedure to select one node as a dictator to propose new blocks. However, this procedure can still lead to potential attacks that make consensus harder to achieve or lead to coordination issues if multiple, competing chains (i.e., forks) are created with the potential that an untruthful fork might be selected. We explore the potential for mechanisms to be used to achieve consensus that are triggered when there is a dispute impeding consensus. Using the feature that nodes stake tokens in proof of stake (POS) protocols, we construct revelation mechanisms in which the unique (subgame perfect) equilibrium involves validating nodes propose truthful blocks using only the information that exists amongst all nodes. We construct operationally and computationally simple mechanisms under both Byzantine Fault Tolerance and a Longest Chain Rule, and discuss their robustness to attacks. Our perspective is that the use of simple mechanisms is an unexplored area of blockchain consensus and has the potential to mitigate known trade-offs and enhance scalability.
△ Less
Submitted 20 June, 2022;
originally announced June 2022.
-
A Solomonic Solution to Ownership Disputes: An Application to Blockchain Front-Running
Authors:
Joshua S. Gans,
Richard Holden
Abstract:
Blockchain front-running involves multiple agents, other than the legitimate agent, claiming a payment from performing a contract. It arises because of the public nature of blockchain transactions and potential network congestion. This paper notes that disputes over payments are similar to classic ownership disputes (such as King Solomon's dilemma). We propose a simultaneous report mechanism that…
▽ More
Blockchain front-running involves multiple agents, other than the legitimate agent, claiming a payment from performing a contract. It arises because of the public nature of blockchain transactions and potential network congestion. This paper notes that disputes over payments are similar to classic ownership disputes (such as King Solomon's dilemma). We propose a simultaneous report mechanism that resolves Solomon's dilemma (using only ordinal preference information) and also eliminates blockchain front-running. In each case, the mechanism relies on threats to remove ownership from all claimants and preferences from the legitimate claimant over allocations to other agents.
△ Less
Submitted 22 February, 2022;
originally announced February 2022.
-
A cross-study analysis of drug response prediction in cancer cell lines
Authors:
Fangfang Xia,
Jonathan Allen,
Prasanna Balaprakash,
Thomas Brettin,
Cristina Garcia-Cardona,
Austin Clyde,
Judith Cohn,
James Doroshow,
Xiaotian Duan,
Veronika Dubinkina,
Yvonne Evrard,
Ya Ju Fan,
Jason Gans,
Stewart He,
Pinyi Lu,
Sergei Maslov,
Alexander Partin,
Maulik Shukla,
Eric Stahlberg,
Justin M. Wozniak,
Hyunseung Yoo,
George Zaki,
Yitan Zhu,
Rick Stevens
Abstract:
To enable personalized cancer treatment, machine learning models have been developed to predict drug response as a function of tumor and drug features. However, most algorithm development efforts have relied on cross validation within a single study to assess model accuracy. While an essential first step, cross validation within a biological data set typically provides an overly optimistic estimat…
▽ More
To enable personalized cancer treatment, machine learning models have been developed to predict drug response as a function of tumor and drug features. However, most algorithm development efforts have relied on cross validation within a single study to assess model accuracy. While an essential first step, cross validation within a biological data set typically provides an overly optimistic estimate of the prediction performance on independent test sets. To provide a more rigorous assessment of model generalizability between different studies, we use machine learning to analyze five publicly available cell line-based data sets: NCI60, CTRP, GDSC, CCLE and gCSI. Based on observed experimental variability across studies, we explore estimates of prediction upper bounds. We report performance results of a variety of machine learning models, with a multitasking deep neural network achieving the best cross-study generalizability. By multiple measures, models trained on CTRP yield the most accurate predictions on the remaining testing data, and gCSI is the most predictable among the cell line data sets included in this study. With these experiments and further simulations on partial data, two lessons emerge: (1) differences in viability assays can limit model generalizability across studies, and (2) drug diversity, more than tumor diversity, is crucial for raising model generalizability in preclinical screening.
△ Less
Submitted 13 August, 2021; v1 submitted 18 April, 2021;
originally announced April 2021.
-
A Public Website for the Automated Assessment and Validation of SARS-CoV-2 Diagnostic PCR Assays
Authors:
Po-E Li,
Adán Myers y Gutiérrez,
Karen Davenport,
Mark Flynn,
Bin Hu,
Chien-Chi Lo,
Elais Player Jackson,
Migun Shakya,
Yan Xu,
Jason Gans,
Patrick S. G. Chain
Abstract:
Summary: Polymerase chain reaction-based assays are the current gold standard for detecting and diagnosing SARS-CoV-2. However, as SARS-CoV-2 mutates, we need to constantly assess whether existing PCR-based assays will continue to detect all known viral strains. To enable the continuous monitoring of SARS-CoV-2 assays, we have developed a web-based assay validation algorithm that checks existing P…
▽ More
Summary: Polymerase chain reaction-based assays are the current gold standard for detecting and diagnosing SARS-CoV-2. However, as SARS-CoV-2 mutates, we need to constantly assess whether existing PCR-based assays will continue to detect all known viral strains. To enable the continuous monitoring of SARS-CoV-2 assays, we have developed a web-based assay validation algorithm that checks existing PCR-based assays against the ever-expanding genome databases for SARS-CoV-2 using both thermodynamic and edit-distance metrics. The assay screening results are displayed as a heatmap, showing the number of mismatches between each detection and each SARS-CoV-2 genome sequence. Using a mismatch threshold to define detection failure, assay performance is summarized with the true positive rate (recall) to simplify assay comparisons. Availability: https://covid19.edgebioinformatics.org/#/assayValidation. Contact: Jason Gans ([email protected]) and Patrick Chain ([email protected])
△ Less
Submitted 8 June, 2020;
originally announced June 2020.
-
More (or Less) Economic Limits of the Blockchain
Authors:
Neil Gandal,
Joshua S. Gans
Abstract:
This paper extends the blockchain sustainability framework of Budish (2018) to consider proof of stake (in addition to proof of work) consensus mechanisms and permissioned (where the number of nodes are fixed) networks. It is demonstrated that an economically sustainable network will involve the same cost regardless of whether it is proof of work or proof of stake although in the later the cost wi…
▽ More
This paper extends the blockchain sustainability framework of Budish (2018) to consider proof of stake (in addition to proof of work) consensus mechanisms and permissioned (where the number of nodes are fixed) networks. It is demonstrated that an economically sustainable network will involve the same cost regardless of whether it is proof of work or proof of stake although in the later the cost will take the form of illiquid financial resources. In addition, it is shown that regulating the number of nodes (as in a permissioned network) does not lead to additional cost savings that cannot otherwise be achieved via a setting of block rewards in a permissionless (i.e., free entry) network. This suggests that permissioned networks will not be able to economize on costs relative to permissionless networks.
△ Less
Submitted 22 December, 2020; v1 submitted 27 November, 2019;
originally announced November 2019.
-
Self-Regulating Artificial General Intelligence
Authors:
Joshua S. Gans
Abstract:
Here we examine the paperclip apocalypse concern for artificial general intelligence (or AGI) whereby a superintelligent AI with a simple goal (ie., producing paperclips) accumulates power so that all resources are devoted towards that simple goal and are unavailable for any other use. We provide conditions under which a paper apocalypse can arise but also show that, under certain architectures fo…
▽ More
Here we examine the paperclip apocalypse concern for artificial general intelligence (or AGI) whereby a superintelligent AI with a simple goal (ie., producing paperclips) accumulates power so that all resources are devoted towards that simple goal and are unavailable for any other use. We provide conditions under which a paper apocalypse can arise but also show that, under certain architectures for recursive self-improvement of AIs, that a paperclip AI may refrain from allowing power capabilities to be developed. The reason is that such developments pose the same control problem for the AI as they do for humans (over AIs) and hence, threaten to deprive it of resources for its primary goal.
△ Less
Submitted 15 February, 2018; v1 submitted 12 November, 2017;
originally announced November 2017.
-
Identification of High $\rm p_{\perp}$ Particles with the STAR-RICH Detector
Authors:
STAR-RICH Collaboration,
:,
A. Braem,
D. Cozza,
M. Davenport,
G. De Cataldo,
L. Dell Olio,
D. DiBari,
A. DiMauro,
J. C. Dunlop,
E. Finch,
D. Fraissard,
A. Franco,
J. Gans,
B. Ghidini,
J. W. Harris,
M. Horsley,
G. J. Kunde,
B. Lasiuk,
Y. Lesenechal,
R. D. Majka,
P. Martinengo,
A. Morsch,
E. Nappi,
G. Paic
, et al. (12 additional authors not shown)
Abstract:
The STAR-RICH detector extends the particle identification capapbilities of the STAR experiment for charged hadrons at mid-rapidity. This detector represents the first use of a proximity-focusing CsI-based RICH detector in a collider experiment. It provides identification of pions and kaons up to 3 GeV/c and protons up to 5 GeV/c. The characteristics and performance of the device in the inaugura…
▽ More
The STAR-RICH detector extends the particle identification capapbilities of the STAR experiment for charged hadrons at mid-rapidity. This detector represents the first use of a proximity-focusing CsI-based RICH detector in a collider experiment. It provides identification of pions and kaons up to 3 GeV/c and protons up to 5 GeV/c. The characteristics and performance of the device in the inaugural RHIC run are described.
△ Less
Submitted 25 November, 2002;
originally announced November 2002.
