检索结果-内蒙古大学图书馆

FEDERATED NEURAL BANDITS

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Dai, Zhongxiang Shu, Yao Verma, Arun Fan, Flint Xiaofeng Low, Bryan Kian Hsiang Jaillet, Patrick Department of Computer Science National University of Singapore Singapore Department of Electrical Engineering and Computer Science MIT United States

Recent works on neural contextual bandits have achieved compelling performances due to their ability to leverage the strong representation power of neural networks (NNs) for reward prediction. Many applications of contextual bandits involve multiple agents who collaborate without sharing raw observations, thus giving rise to the setting of federated contextual bandits. Existing works on federated contextual bandits rely on linear or kernelized bandits, which may fall short when modeling complex real-world reward functions. So, this paper introduces the federated neural-upper confidence bound (FN-UCB) algorithm. To better exploit the federated setting, FN-UCB adopts a weighted combination of two UCBs: UCBa allows every agent to additionally use the observations from the other agents to accelerate exploration (without sharing raw observations), while UCBb uses an NN with aggregated parameters for reward prediction in a similar way to federated averaging for supervised learning. Notably, the weight between the two UCBs required by our theoretical analysis is amenable to an interesting interpretation, which emphasizes UCBa initially for accelerated exploration and relies more on UCBb later after enough observations have been collected to train the NNs for accurate reward prediction (i.e., reliable exploitation). We prove sub-linear upper bounds on both the cumulative regret and the number of communication rounds of FN-UCB, and empirically demonstrate its competitive performance. Copyright © 2022, The Authors. All rights reserved.

关键词： Forecasting

Beyond Barren Plateaus: Quantum Variational Algorithms Are Swamped With Traps

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Anschuetz, Eric R. Kiani, Bobak T. MIT Center for Theoretical Physics United States Department of Electrical Engineering and Computer Science MIT United States

One of the most important properties of classical neural networks is how surprisingly trainable they are, though their training algorithms typically rely on optimizing complicated, nonconvex loss functions. Previous results have shown that unlike the case in classical neural networks, variational quantum models are often not trainable. The most studied phenomenon is the onset of barren plateaus in the training landscape of these quantum models, typically when the models are very deep. This focus on barren plateaus has made the phenomenon almost synonymous with the trainability of quantum models. Here, we show that barren plateaus are only a part of the story. We prove that a wide class of variational quantum models—which are shallow, and exhibit no barren plateaus—have only a superpolynomially small fraction of local minima within any constant energy from the global minimum, rendering these models untrainable if no good initial guess of the optimal parameters is known. We also study the trainability of variational quantum algorithms from a statistical query framework, and show that noisy optimization of a wide variety of quantum models is impossible with a sub-exponential number of queries. Finally, we numerically confirm our results on a variety of problem instances. Though we exclude a wide variety of quantum algorithms here, we give reason for optimism for certain classes of variational algorithms and discuss potential ways forward in showing the practical utility of such algorithms. Copyright © 2022, The Authors. All rights reserved.

关键词： Quantum theory

Machine Learning Detection of Majorana Zero Modes from Zero Bias Peak Measurements

学校读者我要写书评

暂无评论

SSRN

SSRN 2024年

作者： Cheng, Mouyang Okabe, Ryotaro Chotrattanapituk, Abhijatmedhi Li, Mingda Quantum Measurement Group MIT CambridgeMA02139 United States School of Physics Peking University Beijing100084 China Department of Chemistry MIT CambridgeMA02139 United States Department of Electrical Engineering and Computer Science MIT CambridgeMA02139 United States Department of Nuclear Science and Engineering MIT CambridgeMA02139 United States

Majorana zero modes (MZMs), emerging as exotic quasiparticles that carry non-Abelian statistics, hold great promise for achieving fault-tolerant topological quantum computation. A key signature of the presence of MZMs is the zero-bias peaks (ZBPs) from tunneling differential conductance. However, the identification of MZMs from ZBPs has faced tremendous challenges, due to the presence of topological trivial states that generate spurious ZBP signals. In this work, we introduce a machine-learning framework that can discern MZM from other signals using ZBP data. Quantum transport simulation from tight-binding models is used to generate the training data, while persistent cohomology analysis confirms the feasibility of classification via machine learning. In particular, even with added data noise, XGBoost classifier reaches 85% accuracy for 1D tunneling conductance data and 94% for 2D data incorporating Zeeman splitting. Tests on prior ZBP experiments show that some data are more likely to originate from MZM than others. Our model offers a quantitative approach to assess MZMs using ZBP data. Furthermore, our results shed light on the use of machine learning on exotic quantum systems with experimental-computational integration. © 2024, The Authors. All rights reserved.

关键词： Machine learning

The SSL Interplay: Augmentations, Inductive Bias, and Generalization

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Cabannes, Vivien Kiani, Bobak T. Balestriero, Randall LeCun, Yann Bietti, Alberto Meta AI New YorkNY United States MIT Department of Electrical Engineering and Computer Science CambridgeMA United States

Self-supervised learning (SSL) has emerged as a powerful framework to learn representations from raw data without supervision. Yet in practice, engineers face issues such as instability in tuning optimizers and collapse of representations during training. Such challenges motivate the need for a theory to shed light on the complex interplay between the choice of data augmentation, network architecture, and training algorithm. We study such an interplay with a precise analysis of generalization performance on both pretraining and downstream tasks in a theory friendly setup, and highlight several insights for SSL practitioners that arise from our *** Codes 68Q32 © 2023, CC BY-NC-SA.

关键词： Personnel training

Algorithmic Pluralism A Structural Approach To Equal Opportunity

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Jain, Shomik Suriyakumar, Vinith M. Creel, Kathleen A. Wilson, Ashia C. Institute for Data Systems and Society MIT United States Department of Electrical Engineering and Computer Science MIT United States Department of Philosophy and Religion Khoury College of Computer Sciences Northeastern University United States

We present a structural approach toward achieving equal opportunity in systems of algorithmic decision-making called algorithmic pluralism. Algorithmic pluralism describes a state of affairs in which no set of algorithms severely limits access to opportunity, allowing individuals the freedom to pursue a diverse range of life paths. To argue for algorithmic pluralism, we adopt Joseph Fishkin’s theory of bottlenecks, which focuses on the structure of decision-points that determine how opportunities are allocated. The theory contends that each decision-point or "bottleneck" limits access to opportunities with some degree of severity and legitimacy. We adopt Fishkin’s structural viewpoint and use it to reframe existing systemic concerns about equal opportunity in algorithmic decision-making, such as patterned inequality and algorithmic monoculture. In our proposal of algorithmic pluralism, we argue for the urgent priority of alleviating severe bottlenecks. We contend that there must be a pluralism of opportunity available to many different individuals in order to promote equal opportunity in a systemic way. We further show how this framework has several implications for system design and regulation through current debates about equal opportunity in algorithmic *** Codes 68-06 © 2023, CC BY.

关键词： Decision making

Stubborn: A Strong Baseline for Indoor Object Navigation

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Luo, Haokuan Yue, Albert Hong, Zhang-Wei Agrawal, Pulkit Improbable AI Lab The Department Of Electrical Engineering And Computer Science MIT United States

We present a strong baseline that surpasses the performance of previously published methods on the Habitat Challenge task of navigating to a target object in indoor environments. Our method is motivated from primary failure modes of prior state-of-the-art: poor exploration, inaccurate object identification, and agent getting trapped due to imprecise map construction. We make three contributions to mitigate these issues: (i) First, we show that existing map-based methods fail to effectively use semantic clues for exploration. We present a semantic-agnostic exploration strategy (called STUBBORN) without any learning that surprisingly outperforms prior work. (ii) We propose a strategy for integrating temporal information to improve object identification. (iii) Lastly, due to inaccurate depth observation the agent often gets trapped in small regions. We develop a multi-scale collision map for obstacle identification that mitigates this issue. Website: https://***/Improbable-AI/Stubborn. © 2022, CC BY.

关键词： Semantics

Large Language Model-Guided Prediction Toward Quantum Materials Synthesis

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Okabe, Ryotaro West, Zack Chotrattanapituk, Abhijatmedhi Cheng, Mouyang Carrizales, Denisse Córdova Xie, Weiwei Cava, Robert J. Li, Mingda Quantum Measurement Group MIT CambridgeMA United States Department of Chemistry MIT CambridgeMA United States Department of Electrical Engineering and Computer Science MIT CambridgeMA United States Center for Computational Science and Engineering MIT CambridgeMA United States Department of Materials Science and Engineering MIT CambridgeMA United States Department of Nuclear Science and Engineering MIT CambridgeMA United States Department of Chemistry Michigan State University East LansingMI United States Department of Chemistry Princeton University PrincetonNJ United States

The synthesis of inorganic crystalline materials is essential for modern technology, especially in quantum materials development. However, designing efficient synthesis workflows remains a significant challenge due to the precise experimental conditions and extensive trial and error. Here, we present a framework using large language models (LLMs) to predict synthesis pathways for inorganic materials, including quantum materials. Our framework contains three models: LHS2RHS, predicting products from reactants;RHS2LHS, predicting reactants from products;and TGT2CEQ, generating full chem-ical equations for target compounds. Fine-tuned on a text-mined synthesis database, our model raises accuracy from under 40% with pretrained models, to under 80% using conventional fine-tuning, and further to around 90% with our proposed generalized Tanimoto similarity, while maintaining robust to additional synthesis steps. Our model further demonstrates comparable performance across materials with varying degrees of quantumness quantified using quantum weight, indicating that LLMs offer a powerful tool to predict balanced chemical equations for quantum materials discovery. © 2024, CC BY.

关键词： Nanocrystals

Continuous Deep Equilibrium Models: Training Neural ODEs faster by integrating them to Infinity

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Pal, Avik Edelman, Alan Rackauckas, Christopher CSAIL MIT United States Department of Electrical Engineering and Computer Science MIT United States Department of Mathematics MIT United States

Implicit models separate the definition of a layer from the description of its solution process. While implicit layers allow features such as depth to adapt to new scenarios and inputs automatically, this adaptivity makes its computational expense challenging to predict. In this manuscript, we increase the "implicitness" of the DEQ by redefining the method in terms of an infinite time neural ODE, which paradoxically decreases the training cost over a standard neural ODE by 2−4×. Additionally, we address the question: is there a way to simultaneously achieve the robustness of implicit layers while allowing the reduced computational expense of an explicit layer? To solve this, we develop Skip and Skip Reg. DEQ, an implicit-explicit (IMEX) layer that simultaneously trains an explicit prediction followed by an implicit correction. We show that training this explicit predictor is free and even decreases the training time by 1.11 − 3.19×. Together, this manuscript shows how bridging the dichotomy of implicit and explicit deep learning can combine the advantages of both techniques. Copyright © 2022, The Authors. All rights reserved.

关键词： Ordinary differential equations

Sample-then-optimize batch neural thompson sampling 22

学校读者我要写书评

暂无评论

Sample-then-optimize batch neural thompson sampling

Proceedings of the 36th International Conference on Neural Information Processing Systems

作者： Zhongxiang Dai Yao Shu Bryan Kian Hsiang Low Patrick Jaillet Dept. of Computer Science National University of Singapore Republic of Singapore Dept. of Electrical Engineering and Computer Science MIT

ISBN: (纸本)9781713871088

Bayesian optimization (BO), which uses a Gaussian process (GP) as a surrogate to model its objective function, is popular for black-box optimization. However, due to the limitations of GPs, BO underperforms in some problems such as those with categorical, high-dimensional or image inputs. To this end, recent works have used the highly expressive neural networks (NNs) as the surrogate model and derived theoretical guarantees using the theory of neural tangent kernel (NTK). However, these works suffer from the limitations of the requirement to invert an extremely large parameter matrix and the restriction to the sequential (rather than batch) setting. To overcome these limitations, we introduce two algorithms based on the Thompson sampling (TS) policy named Sample-Then-Optimize Batch Neural TS (STO-BNTS) and STO-BNTS-Linear. To choose an input query, we only need to train an NN (resp. a linear model) and then choose the query by maximizing the trained NN (resp. linear model), which is equivalently sampled from the GP posterior with the NTK as the kernel function. As a result, our algorithms sidestep the need to invert the large parameter matrix yet still preserve the validity of the TS policy. Next, we derive regret upper bounds for our algorithms with batch evaluations, and use insights from batch BO and NTK to show that they are asymptotically no-regret under certain conditions. Finally, we verify their empirical effectiveness using practical AutoML and reinforcement learning experiments.

关键词：