An official website of the United States government
We study the problem of observing quantum collective phenomena emerging from large numbers of measurements. These phenomena are difficult to observe in conventional experiments because, in order to distinguish the effects of measurement from dephasing, it is necessary to postselect on sets of measurement outcomes with Born probabilities that are exponentially small in the number of measurements performed. An unconventional approach, which avoids this exponential “postselection problem”, is to construct cross-correlations between experimental data and the results of simulations on classical computers. However, these cross-correlations generally have no definite relation to physical quantities. We first show how to incorporate classical shadows into this framework, thereby allowing for the construction of quantum information-theoretic cross-correlations. We then identify cross-correlations that both upper and lower bound the measurement-averaged von Neumann entanglement entropy, as well as cross-correlations that lower bound the measurement-averaged purity and entanglement negativity. These bounds show that experiments can be performed to constrain postmeasurement entanglement without the need for postselection. To illustrate our technique, we consider how it could be used to observe the measurement-induced entanglement transition in Haar-random quantum circuits. We use exact numerical calculations as proxies for quantum simulations and, to highlight the fundamental limitations of classical memory, we construct cross-correlations with tensor-network calculations at finite bond dimension. Our results reveal a signature of measurement-induced criticality that can be observed using a quantum simulator in polynomial time and with polynomial classical memory. Published by the American Physical Society2024
more »
« less
Sequential learning on a tensor network Born machine with trainable token embedding
Abstract Generative models aim to learn the probability distributions underlying data, enabling the generation of new, realistic samples. Quantum-inspired generative models, such as Born machines based on the matrix product state (MPS) framework, have demonstrated remarkable capabilities in unsupervised learning tasks. This study advances the Born machine paradigm by introducing trainable token embeddings through positive operator-valued measurements (POVMs), replacing the traditional approach of static tensor indices. Key technical innovations include encoding tokens as quantum measurement operators with trainable parameters and leveraging QR decomposition to adjust the physical dimensions of the MPS. This approach maximizes the utilization of operator space and enhances the model’s expressiveness. Empirical results on RNA data demonstrate that the proposed method significantly reduces negative log-likelihood compared to one-hot embeddings, with higher physical dimensions further enhancing single-site probabilities and multi-site correlations. The model also outperforms GPT-2 in single-site estimation and achieves competitive correlation modeling, showcasing the potential of trainable POVM embeddings for complex data correlations in quantum-inspired sequence modeling.
more »
« less
- Award ID(s):
- 2238360
- PAR ID:
- 10610057
- Publisher / Repository:
- IOP Publishing
- Date Published:
- Journal Name:
- Machine Learning: Science and Technology
- Volume:
- 6
- Issue:
- 2
- ISSN:
- 2632-2153
- Format(s):
- Medium: X Size: Article No. 025071
- Size(s):
- Article No. 025071
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract Despite the growing use of Aquatic Ecosystem Models for lake modeling, there is currently no widely applicable framework for their configuration, calibration, and evaluation. Calibration is generally based on direct data comparison of observed versus modeled state variables using standard statistical techniques, however, this approach may not give a complete picture of the model's ability to capture system‐scale behavior that is not easily perceivable in observations, but which may be important for resource management. The aim of this study is to compare the performance of “naïve” calibration and a “system‐inspired” calibration, an approach that augments the standard state‐based calibration with a range of system‐inspired metrics (e.g., thermocline depth, metalimnetic oxygen minima), to increase the coherence between the simulated and natural ecosystems. A coupled physical‐biogeochemical model was applied to a focal site to simulate two key state‐variables: water temperature and dissolved oxygen. The model was calibrated according to the new system‐inspired modeling convention, using formal calibration techniques. There was an improvement in the simulation using parameters optimized on the additional metrics, which helped to reduce uncertainty predicting aspects of the system relevant to reservoir management, such as the occurrence of the metalimnetic oxygen minima. Extending the use of system‐inspired metrics when calibrating models has the potential to improve model fidelity for capturing more complex ecosystem dynamics.more » « less
-
A wide variety of machine learning tasks such as knowledge base completion, ontology alignment, and multi-label classification can benefit from incorporating into learning differentiable representations of graphs or taxonomies. While vectors in Euclidean space can theoretically represent any graph, much recent work shows that alternatives such as complex, hyperbolic, order, or box embeddings have geometric properties better suited to modeling real-world graphs. Experimentally these gains are seen only in lower dimensions, however, with performance benefits diminishing in higher dimensions. In this work, we introduce a novel variant of box embeddings that uses a learned smoothing parameter to achieve better representational capacity than vector models in low dimensions, while also avoiding performance saturation common to other geometric models in high dimensions. Further, we present theoretical results that prove box embeddings can represent any DAG. We perform rigorous empirical evaluations of vector, hyperbolic, and region-based geometric representations on several families of synthetic and real-world directed graphs. Analysis of these results exposes correlations between different families of graphs, graph characteristics, model size, and embedding geometry, providing useful insights into the inductive biases of various differentiable graph representations.more » « less
-
Abstract 1. Species distribution models (SDMs) are crucial tools for understanding and predicting biodiversity patterns, yet they often struggle with limited data, biased sampling, and complex species-environment relationships. Here I present NicheFlow, a novel foundation model for SDMs that leverages generative AI to address these challenges and advance our ability to model and predict species distributions across taxa and environments. 2. NicheFlow employs a two-stage generative approach, combining species embeddings with two chained generative models, one to generate a distribution in environmental space, and a second to generate a distribution in geographic space. This architecture allows for the sharing of information across species and captures complex, non-linear relationships in environmental space. I trained NicheFlow on a comprehensive dataset of reptile distributions and evaluated its performance using both standard SDM metrics and zero-shot prediction tasks. 3. NicheFlow demonstrates good predictive performance, particularly for rare and data-deficient species. The model successfully generated plausible distributions for species not seen during training, showcasing its potential for zero-shot prediction. The learned species embeddings captured meaningful ecological information, revealing patterns in niche structure across taxa, latitude and range sizes. 4. As a proof-of-principle foundation model, NicheFlow represents a significant advance in species distribution modeling, offering a powerful tool for addressing pressing questions in ecology, evolution, and conservation biology. Its ability to model joint species distributions and generate hypothetical niches opens new avenues for exploring ecological and evolutionary questions, including ancestral niche reconstruction and community assembly processes. This approach has the potential to transform our understanding of biodiversity patterns and improve our capacity to predict and manage species distributions in the face of global change.more » « less
-
Quantum generative models hold the promise of accelerating or improving machine learning tasks by leveraging the probabilistic nature of quantum states, but the successful optimization of these models remains a difficult challenge. To tackle this challenge, we present a new architecture for quantum generative modeling that combines insights from classical machine learning and quantum phases of matter. In particular, our model utilizes both many-body localized (MBL) dynamics and hidden units to improve the optimization of the model. We demonstrate the applicability of our model on a diverse set of classical and quantum tasks, including a toy version of MNIST handwritten digits, quantum data obtained from quantum many-body states, and nonlocal parity data. Our architecture and algorithm provide novel strategies of utilizing quantum many-body systems as learning resources and reveal a powerful connection between disorder, interaction, and learning in quantum many-body systems. Published by the American Physical Society2024more » « less
