检索结果-内蒙古大学图书馆

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Kadıoğlu, Serdar Zhu, Elton Yechao Rosenberg, Gili Brubaker, John Kyle Schuetz, Martin J.A. Salton, Grant Zhu, Zhihuai Katzgraber, Helmut G. AI Center of Excellence Fidelity Investments MA United States Department of Computer Science Brown University RI United States Fidelity Center for Applied Technology FMR LLC MA United States Amazon Quantum Solutions Lab WA United States AWS Center for Quantum Computing CA United States Institute for Quantum Information and Matter California Institute of Technology CA United States

ISBN: (纸本)157735897X

In this paper, we design, develop, and release BoolXAI, an interpretable machine learning classification approach for Explainable AI (XAI) based on expressive Boolean formulas. The Boolean formula defines a logical rule with tunable complexity according to which input data are classified. Beyond the classical conjunction and disjunction, BoolXAI offers expressive operators such as AtLeast, AtMost, and Choose and their parameterization. This provides higher expressiveness compared to rigid rules- and tree-based approaches. We show how to train BoolXAI classifiers effectively using native local optimization to search the space of feasible formulas. We provide illustrative results on several well-known public benchmarks that demonstrate the competitive nature of our approach compared to existing methods. Our work is embodied in the open-source BoolXAI library with a high-level user interface to serve researchers and practitioners. BoolXAI can be used either as a standalone interpretable classifier or for post-hoc explanations of other black-box models or observed behavior. We highlight several desirable benefits of our tool, especially in industrial settings where rapid experimentation, reusability, reproducibility, deployment, and maintenance are of great interest. Finally, we showcase a deployed service powered by BoolXAI as an enterprise application. Copyright © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Boolean functions

来源：评论

学校读者我要写书评

暂无评论

PROF: Protected Order Flow in a Profit-Seeking World

arXiv

引用

arXiv 2024年

作者： Babel, Kushal Jean-Louis, Nerla Ji, Yan Misra, Ujval Kelkar, Mahimna Mudiyanselage, Kosala Yapa Miller, Andrew Juels, Ari Cornell Tech United States UIUC United States UC Berkeley United States IC3 Fidelity Center for Applied Technology

Users of decentralized finance (DeFi) applications face significant risks from adversarial actions that manipulate the order of transactions to extract value from users. Such actions—an adversarial form of what is called maximal-extractable value (MEV)—impact both individual outcomes and the stability of the DeFi ecosystem. MEV exploitation, moreover, is being institutionalized through an architectural paradigm known Proposer-Builder Separation (PBS). This work introduces a system called PROF (Protected Order Flow) that is designed to limit harmful forms of MEV in existing PBS systems. PROF aims at this goal using two ideas. First, PROF imposes an ordering on a set ("bundle") of privately input transactions and enforces that ordering all the way through to block production—preventing transaction-order manipulation. Second, PROF creates bundles whose inclusion is profitable to block producers, thereby ensuring that bundles see timely inclusion in blocks. PROF is backward-compatible, meaning that it works with existing and future PBS designs. PROF is also compatible with any desired algorithm for ordering transactions within a PROF bundle (e.g., first-come, first-serve, fee-based, etc.). It executes efficiently, i.e., with low latency, and requires no additional trust assumptions among PBS entities. We quantitatively and qualitatively analyze PROF’s incentive structure, and its utility to users compared with existing solutions. We also report on inclusion likelihood of PROF transactions, and concrete latency numbers through our end-to-end implementation. © 2024, CC BY.

关键词： Decentralized finance

来源：评论

学校读者我要写书评

暂无评论

Solving QUBOs with a quantum-amenable branch and bound method

arXiv

引用

arXiv 2024年

作者： Häner, Thomas Booth, Kyle E.C. Borujeni, Sima E. Zhu, Elton Yechao Amazon Quantum Solutions Lab SeattleWA98170 United States Fidelity Center for Applied Technology FMR LLC BostonMA02210 United States

Due to the expected disparity in quantum vs. classical clock speeds, quantum advantage for branch and bound algorithms is more likely achievable in settings involving large search trees and low operator evaluation costs. Therefore, in this paper, we describe and experimentally validate an exact classical branch and bound solver for quadratic unconstrained binary optimization (QUBO) problems that matches these criteria. Our solver leverages cheap-to-implement bounds from the literature previously proposed for Ising models, including that of Hartwig, Daske, and Kobe from 1984. We detail a variety of techniques from high-performance computing and operations research used to boost solver performance, including a global variable reordering heuristic, a primal heuristic based on simulated annealing, and a truncated computation of the recursive bound. We also outline a number of simple and inexpensive bound extrapolation techniques. Finally, we conduct an extensive empirical analysis of our solver, comparing its performance to state-of-the-art QUBO and MaxCut solvers, and discuss the challenges of a speedup via quantum branch and bound beyond those faced by any quadratic quantum speedup. Copyright © 2024, The Authors. All rights reserved.

关键词： Simulated annealing

来源：评论

学校读者我要写书评

暂无评论

Quantum State Preparation of Normal Distributions using Matrix Product States

arXiv

引用

arXiv 2023年

作者： Iaconis, Jason Johri, Sonika Zhu, Elton Yechao IonQ Inc 4505 Campus Dr College ParkMD20740 United States Fidelity Center for Applied Technology FMR LLC BostonMA02210 United States

State preparation is a necessary component of many quantum algorithms. In this work, we combine a method for efficiently representing smooth differentiable probability distributions using matrix product states with recently discovered techniques for initializing quantum states to approximate matrix product states. Using this, we generate quantum states encoding a class of normal probability distributions in a trapped ion quantum computer for up to 20 qubits. We provide an in depth analysis of the different sources of error which contribute to the overall fidelity of this state preparation procedure. Our work provides a study in quantum hardware for scalable distribution loading, which is the basis of a wide range of algorithms that provide quantum advantage. © 2023, CC BY.

关键词： Normal distribution

来源：评论

学校读者我要写书评

暂无评论

Generative quantum learning of joint probability distribution functions

引用

Physical Review Research 2022年第4期4卷 043092-043092页

作者： Elton Yechao Zhu Sonika Johri Dave Bacon Mert Esencan Jungsang Kim Mark Muir Nikhil Murgai Jason Nguyen Neal Pisenti Adam Schouela Ksenia Sosnova Ken Wright Fidelity Center for Applied Technology FMR LLC Boston Massachusetts 02210 USA IonQ Inc 4505 Campus Dr College Park Maryland 20740 USA

Modeling joint probability distributions is an important task in a wide variety of fields. One popular technique for this employs a family of multivariate distributions with uniform marginals called copulas. While the theory of modeling joint distributions via copulas is well understood, it gets practically challenging to accurately model real data with many variables. In this paper, we show that any copula can be naturally mapped to a multipartite maximally entangled state. Thus, the task of learning joint probability distributions becomes the task of learning maximally entangled states. We prove that a variational ansatz we christen as a “qopula” based on this insight leads to an exponential advantage over classical methods of learning some joint distributions. As an application, we train a quantum generative adversarial network (QGAN) and a quantum circuit Born machine (QCBM) using this variational ansatz to generate samples from joint distributions of two variables in historical data from the stock market. We demonstrate our generative learning algorithms on trapped ion quantum computers from IonQ for up to eight qubits. Our experimental results show interesting findings such as the resilience against noise, outperformance against equivalent classical models and 20–1000 times less iterations required to converge as compared to equivalent classical models.

关键词： Machine learning Quantum algorithms

来源：评论

学校读者我要写书评

暂无评论

Scalable iterative pruning of large language and vision models using block coordinate descent

arXiv

引用

arXiv 2024年

作者： Rosenberg, Gili Brubaker, J. Kyle Schuetz, Martin J.A. Zhu, Elton Yechao Kadıoğlu, Serdar Borujeni, Sima E. Katzgraber, Helmut G. Amazon Quantum Solutions Lab SeattleWA98170 United States AWS Center for Quantum Computing PasadenaCA91125 United States Fidelity Center for Applied Technology FMR LLC BostonMA02210 United States AI Center of Excellence FMR LLC BostonMA02210 United States

Pruning neural networks, which involves removing a fraction of their weights, can often maintain high accuracy while significantly reducing model complexity, at least up to a certain limit. We present a neural network pruning technique that builds upon the Combinatorial Brain Surgeon, but solves an optimization problem over a subset of the network weights in an iterative, block-wise manner using block coordinate descent. The iterative, block-based nature of this pruning technique, which we dub "iterative Combinatorial Brain Surgeon" (iCBS) allows for scalability to very large models, including large language models (LLMs), that may not be feasible with a one-shot combinatorial optimization approach. When applied to large models like Mistral and DeiT, iCBS achieves higher performance metrics at the same density levels compared to existing pruning methods such as Wanda. This demonstrates the effectiveness of this iterative, block-wise pruning method in compressing and optimizing the performance of large deep learning models, even while optimizing over only a small fraction of the weights. Moreover, our approach allows for a quality-time (or cost) tradeoff that is not available when using a one-shot pruning technique alone. The block-wise formulation of the optimization problem enables the use of hardware accelerators, potentially offsetting the increased computational costs compared to one-shot pruning methods like Wanda. In particular, the optimization problem solved for each block is quantum-amenable in that it could, in principle, be solved by a quantum computer. Copyright © 2024, The Authors. All rights reserved.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

Explainable AI using expressive Boolean formulas

arXiv

引用

arXiv 2023年

作者： Rosenberg, Gili Brubaker, J. Kyle Schuetz, Martin J.A. Salton, Grant Zhu, Zhihuai Zhu, Elton Yechao Kadıoğlu, Serdar Borujeni, Sima E. Katzgraber, Helmut G. Amazon Quantum Solutions Lab SeattleWA98170 United States AWS Center for Quantum Computing PasadenaCA91125 United States California Institute of Technology PasadenaCA91125 United States Fidelity Center for Applied Technology FMR LLC BostonMA02210 United States AI Center of Excellence FMR LLC BostonMA02210 United States

We propose and implement an interpretable machine learning classification model for Explainable AI (XAI) based on expressive Boolean formulas. Potential applications include credit scoring and diagnosis of medical conditions. The Boolean formula defines a rule with tunable complexity (or interpretability), according to which input data are classified. Such a formula can include any operator that can be applied to one or more Boolean variables, thus providing higher expressivity compared to more rigid rule-based and tree-based approaches. The classifier is trained using native local optimization techniques, efficiently searching the space of feasible formulas. Shallow rules can be determined by fast Integer Linear Programming (ILP) or Quadratic Unconstrained Binary Optimization (QUBO) solvers, potentially powered by special purpose hardware or quantum devices. We combine the expressivity and efficiency of the native local optimizer with the fast operation of these devices by executing non-local moves that optimize over subtrees of the full Boolean formula. We provide extensive numerical benchmarking results featuring several baselines on well-known public datasets. Based on the results, we find that the native local rule classifier is generally competitive with the other classifiers. The addition of non-local moves achieves similar results with fewer iterations, and therefore using specialized or quantum hardware could lead to a speedup by fast proposal of non-local moves. Copyright © 2023, The Authors. All rights reserved.

关键词： Integer programming

来源：评论

学校读者我要写书评

暂无评论

Custody protocols using bitcoin vaults

arXiv

引用

arXiv 2020年

作者： Swambo, Jacob Hommel, Spencer McElrath, Bob Bishop, Bryan King's College London Department of Informatics Fidelity Center for Applied Technology

A bitcoin covenant is a mechanism to enforce conditions on future bitcoin transactions. A bitcoin vault is a specific type of covenant transaction that enforces a time-lock on the transfer of control of funds to a hot wallet, but enables an immediate transfer of funds into a deep cold recovery wallet. This paper demonstrates how to integrate a bitcoin vault into a custody protocol and demonstrates the security properties of that protocol. The vault is implemented using pre-signed transactions with secure key deletion (as proposed in [SHMB20]). It is shown that vault-custody protocols enable the wallet owner to specify their desired balance for an inherent trade-off between the security of and accessibility of bitcoin holdings by adjusting the length of time-locks used. It is also demonstrated that wallet owners have increased control of risk-management by compartmentalizing funds across numerous vault transactions. While it isn't realistic to completely prevent theft, the most likely theft scenarios (compromising the hot wallet) have severely limited profitability for an attacker, deterring attempts at theft from the beginning. The proposed architecture was designed to offer defence-in-depth through redundancy and fault-tolerant functionality as well as countermeasures for class breaks through diversity across hardware and software layers. Finally, the architecture employs a detection (a watchtower) and response system that enables fail-safe recovery from attempted or partial thefts through a second type of covenant transaction, a push-to-recovery-wallet transaction. Copyright © 2020, The Authors. All rights reserved.

关键词： Crime

来源：评论

学校读者我要写书评

暂无评论

Bitcoin covenants: Three ways to control the future

arXiv

引用

arXiv 2020年

作者： Swambo, Jacob Hommel, Spencer McElrath, Bob Bishop, Bryan King’s College London Department of Informatics Fidelity Center for Applied Technology

A bitcoin covenant is a mechanism to enforce conditions on how the control of coins will be transferred in the future. This work introduces a mechanism to construct a general class of covenants without requiring a change to the consensus rules of bitcoin, in contrast to previous covenant mechanism proposals. An exploration of the broad design space of deleted-key covenants (using pre-signed transactions with secure key deletion) is given with security analyses that demonstrate a range of possible security models. While the power derived from a mechanism for covenants is undisputed, the method of implementation is contentious. One purpose of this work is to contribute to that debate by demonstrating what is possible today without introducing new security risks to bitcoin. On the other hand, this work makes a compelling case for what can be gained through a soft-fork upgrade for either a Script-based covenant mechanism [Rub20] or a change to the signature hash (SIGHASH) system [Dec17]. The former has had several approaches proposed previously. The latter, introduced herein, is based on an independent proposal aimed at improving off-chain protocols which can be exapted to enable recovered-key covenants through elliptic curve key recovery. The dominant differing factor between a covenant mechanism that uses pre-signed transactions with secure key deletion and those that require soft-fork upgrades is the basis of security for each method. The process of secure key deletion is subject to a trade-off between convenience and security. Secure key deletion requires more procedural overhead for a higher level of security (e.g. by relying on multi-signature pre-signed transactions with more signing keys to add redundancy to the deletion process). Moreover, it is recommended that the deletion occurs quickly, which in a multi-party context necessitates interactivity, a practical downside when compared with other mechanisms. Script-based and recovered-key covenants have a non-inte

关键词： Cryptography

来源：评论

学校读者我要写书评

暂无评论

Generative quantum learning of joint probability distribution functions

arXiv

引用

arXiv 2021年

作者： Zhu, Elton Yechao Johri, Sonika Bacon, Dave Esencan, Mert Kim, Jungsang Muir, Mark Murgai, Nikhil Nguyen, Jason Pisenti, Neal Schouela, Adam Sosnova, Ksenia Wright, Ken Fidelity Center for Applied Technology FMR LLC BostonMA02210 United States IonQ Inc 4505 Campus Dr College ParkMD20740 United States

Modeling joint probability distributions is an important task in a wide variety of fields. One popular technique for this employs a family of multivariate distributions with uniform marginals called copulas. While the theory of modeling joint distributions via copulas is well understood, it gets practically challenging to accurately model real data with many variables. In this paper, we show that any copula can be naturally mapped to a multipartite maximally entangled state. Thus, the task of learning joint probability distributions becomes the task of learning maximally entangled states. We prove that a variational ansatz we christen as a 'qopula' based on this insight leads to an exponential advantage over classical methods of learning some joint distributions. As an application, we train a quantum generative adversarial network (QGAN) and a quantum circuit born machine (QCBM) using this variational ansatz to generate samples from joint distributions of two variables in historical data from the stock market. We demonstrate our generative learning algorithms on trapped ion quantum computers from IonQ for up to eight qubits. Our experimental results show interesting findings such as the resilience against noise, outperformance against equivalent classical models and 20-1000 times less iterations required to converge as compared to equivalent classical models. © 2021, CC BY.

关键词： Distribution functions

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：