26838 research outputs found
Sort by
trl-lab/contextual-sensitive-data
No full textThis dataset includes tables with sensitivity annotations that was used to train and evaluate methods for detecting contextual sensitive data
The computation of generalized embeddings for underwater acoustic target recognition using contrastive learning
Full text linkThe increasing level of sound pollution in marine environments poses an increased threat to ocean health, making it crucial to monitor underwater noise. By monitoring this noise, the sources responsible for this pollution can be mapped. Monitoring is performed by passively listening to these sounds. This generates a large amount of data records, capturing a mix of sound sources such as ship activities and marine mammal vocalizations. Although machine learning offers a promising solution for automatic sound classification, current state-of-the-art methods implement supervised learning. This requires a large amount of high-quality labeled data that is not publicly available. In contrast, a massive amount of lower-quality unlabeled data is publicly available, offering the opportunity to explore unsupervised learning techniques. This research explores this possibility by implementing an unsupervised Contrastive Learning approach. Here, a Conformer-based encoder is optimized by the so-called Variance-Invariance-Covariance Regularization loss function on these lower-quality unlabeled data and the translation to the labeled data is made. Through classification tasks involving recognizing ship types and marine mammal vocalizations, our method demonstrates the ability to produce robust and generalized embeddings. This shows the potential of unsupervised methods for various automatic underwater acoustic analysis tasks
Testing and learning structured quantum Hamiltonians
Full text linkWe consider the problems of testing and learning an -qubit Hamiltonian expressed in its Pauli basis, from queries to its evolution operator . To this end, we prove the following results.
1. Testing: We give a tolerant testing protocol to decide if a Hamiltonian is -close to -local or -far from -local in the norm of the coefficients, with queries, thereby solving two open questions posed in a recent work by Bluhm, Caro and Oufkir (Bluhm, A., Caro, M.C., Oufkir, A.). We give a protocol for testing whether a Hamiltonian is -close to being -sparse or -far from being -sparse in the norm of the coefficients, with queries.
2. Learning: We give a protocol to -learn unstructured Hamiltonian in the norm of the coefficients with queries. Combining this with the non-commutative Bohnenblust-Hille inequality, we obtain an algorithm for learning -local Hamiltonians in norm of the coefficients that only uses queries. For Hamiltonians that are -sparse in the Pauli basis, we can learn them in the norm with queries.
3. Learning without quantum memory: The learning results stated above have no dependence on the system size , but require -qubit quantum memory. We give subroutines that allow us to reproduce all the above learning results without quantum memory; squaring the query complexity and paying a -factor in the local case and an -factor in the sparse case.
4. Testing without quantum memory: We give a new subroutine called Pauli hashing, which allows one to tolerantly test -sparse Hamiltonians in the norm using query complexity. A key ingredient is showing that -sparse Pauli channels can be tested in a tolerant fashion as being -close to being -sparse or -far under the diamond norm, using queries via Pauli hashing.
In order to prove these results, we prove new structural theorems for local Hamiltonians, sparse Pauli channels and sparse Hamiltonians. We complement our learning algorithms with lower bounds that are polynomially weaker. Furthermore, our algorithms use short time evolutions and do not assume prior knowledge of the terms on which the Pauli spectrum is supported on, i.e., we do not require prior knowledge of the support of the Hamiltonian
Optimal type-dependent liquid welfare guarantees for autobidding agents with budgets
Full text linkOnline advertising systems have recently transitioned to autobidding, allowing advertisers todelegate bidding decisions to automated agents. Each advertiser directs their agent to optimize an objectivefunction subject to return-on-investment (ROI) and budget constraints. Given their practical relevance, this shifthas spurred a surge of research on the liquid welfare price of anarchy (POA) of fundamental auction formatsunder autobidding, most notably simultaneous first-price auctions (FPA). One of the main challenges is tounderstand the efficiency of FPA in the presence of heterogeneous agent types. We introduce a type-dependentsmoothness framework that enables a unified analysis of the POA in such complex autobidding environments. Inour approach, we derive type-dependent smoothness parameters which we carefully balance to obtain POA bounds.This balancing gives rise to a POA-revealing mathematical program, which we use to determine tight bounds onthe POA of coarse correlated equilibria (CCE). Our framework is versatile enough to handle heterogeneous agenttypes and extends to the general class of fractionally subadditive valuations. Additionally, we develop a novelreduction technique that transforms budget-constrained agents into budget-unconstrained ones. Combining thisreduction technique with our smoothness framework enables us to derive tight bounds on the POA of CCE inthe general hybrid agent model with both ROI and budget constraints. Among other results, our bounds uncoveran intriguing threshold phenomenon showing that the POA depends intricately on the smallest and largest agenttypes. We also extend our study to FPAs with reserve prices, which can be interpreted as predictions of agents’values, to further improve efficiency guarantees
Transformer-based few-shot learning for modeling Electricity Consumption Profiles with minimal data across thousands of domains
Full text linkElectricity Consumption Profiles (ECPs) are crucial for operating and planning power distribution systems, especially with the increasing number of low-carbon technologies such as solar panels and electric vehicles. Traditional ECP modeling methods typically assume the availability of sufficient ECP data. However, in practice, the accessibility of ECP data is limited due to privacy issues or the absence of metering devices. Few-shot learning (FSL) has emerged as a promising solution for ECP modeling in data-scarce scenarios. Nevertheless, standard FSL methods, such as those used for images, are unsuitable for ECP modeling because (1) these methods usually assume several source domains with sufficient data and several target domains. However, in the context of ECP modeling, there may be thousands of source domains, e.g., households with a moderate amount of data, and thousands of target domains, e.g., households that ECP are required to be modeled. (2) Standard FSL methods usually involve cumbersome knowledge transfer mechanisms, such as pre-training and fine-tuning. To address these limitations, this paper proposes a novel FSL framework that integrates Transformers with Gaussian Mixture Models (GMMs) for ECP modeling. The proposed approach is fine-tuning-free, computationally efficient, and robust even with extremely limited data. Results show that our method can accurately restore the complex ECP distribution with a minimal amount of ECP data (e.g., only 1.6% of the complete domain dataset) and outperforms state-of-the-art time series modeling methods in the context of ECP modeling
Assessing fault-tolerant quantum advantage for k-SAT with structure
Full text linkFor many problems, quantum algorithms promise speedups over their classical counterparts. However, these results predominantly rely on asymptotic worst-case analysis, which overlooks significant overheads due to error correction and the fact that real-world instances often contain exploitable structure. In this work, we employ the hybrid benchmarking method to evaluate the potential of quantum Backtracking and Grover’s algorithm against the 2023 SAT competition main track winner in solving random k-SAT instances with tunable structure, designed to represent industry-like scenarios, using both T-depth and T-count as cost metrics to estimate quantum run times. Our findings reproduce the results of Campbell, Khurana, and Montanaro (Quantum ’19) in the unstructured case using hybrid benchmarking. However, we offer a more sobering perspective in practically relevant regimes: almost all quantum speedups vanish, even asymptotically, when minimal structure is introduced or when T-count is considered instead of T-depth. Moreover, when the requirement is for the algorithm to find a solution within a single day, we find that only Grover’s algorithm has the potential to outperform classical algorithms, but only in a very limited regime and only when using T-depth. We also discuss how more sophisticated heuristics could restore the asymptotic scaling advantage for quantum backtracking, but our findings suggest that the potential for practical quantum speedups in more structured k-SAT solving will remain limited
Understanding trust toward human versus AI-generated health information through behavioral and physiological sensing
Full text linkAs AI-generated health information proliferates online and becomes increasingly indistinguishable from human-sourced information, it becomes critical to understand how people trust and label such content, especially when the information is inaccurate. We conducted two complementary studies: (1) a mixed-methods survey (N=142) employing a 2 (source: Human vs. LLM) × 2 (label: Human vs. AI) × 3 (type: General, Symptom, Treatment) design, and (2) a within-subjects lab study (N=40) incorporating eye-tracking and physiological sensing (ECG, EDA, skin temperature). Participants were presented with health information varying by source-label combinations and asked to rate their trust, while their gaze behavior and physiological signals were recorded. We found that LLM-generated information was trusted more than human-generated content, whereas information labeled as human was trusted more than that labeled as AI. Trust remained consistent across information types. Eye-tracking and physiological responses varied significantly by source and label. Machine learning models trained on these behavioral and physiological features predicted binary self-reported trust levels with 73 % accuracy and information source with 65 % accuracy. Our findings demonstrate that adding transparency labels to online health information modulates trust. Behavioral and physiological features show potential to verify trust perceptions and indicate if additional transparency is needed
Moment-sos and spectral hierarchies for polynomial optimization on the sphere and quantum de Finetti theorems
No full textWe revisit the convergence analysis of two approximation hierarchies for polynomial optimization on the unit sphere. The first one is based on the moment-sos approach and gives semidefinite bounds for which Fang and Fawzi (2021) showed an analysis in for the th level bound, using the polynomial kernel method. The second hierarchy was recently proposed by Lovitz and Johnston (2023) and gives spectral bounds for which they show a convergence rate in , using a quantum de Finetti theorem of Christandl et al. (2007) that applies to complex Hermitian matrices with a “double” symmetry. We investigate links between these approaches, in particular, via duality of moments and sums of squares.
Our main results include showing that the spectral bounds cannot have a convergence rate better than and that they do not enjoy generic finite convergence. In addition, we propose alternative performance analyses that involve explicit constants depending on intrinsic parameters of the optimization problem. For this we develop a novel “banded” real de Finetti theorem that applies to real matrices with “double” symmetry. We also show how to use the polynomial kernel method to obtain a de Finetti type result in for real maximally symmetric matrices, improving an earlier result in of Doherty and Wehner (2012)
On admissibility in post-hoc hypothesis testing
No full textThe validity of classical hypothesis testing requires the significance level α be fixed before any statistical analysis takes place. This is a stringent requirement. For instance, it prohibits updating α during (or after) an experiment due to changing concern about the cost of false positives, or to reflect unexpectedly strong evidence against the null. Perhaps most disturbingly, witnessing a p-value p ≪ α vs p=α−ϵ for tiny ϵ ' 0 has no (statistical) relevance for any downstream decision-making. Following recent work of Grünwald [1], we develop a theory of post-hoc hypothesis testing, enabling α to be chosen after seeing and analyzing the data. To study “good” post-hoc tests we introduce Γ-admissibility, where Γ is a set of adversaries which map the data to a significance level. We classify the set of Γ-admissible rules for various sets Γ, showing they must be based on e-values, and recover the Neyman-Pearson lemma when Γ is the constant map