Search

We introduce a new framework linking group theory and formal language theory which generalizes a number of ways these topics have been linked in the past. For a language class C in the Chomsky hierarchy, we say a group is epiC if it admits a language $L$ over a finite (monoidal) generating set $X \subseteq G$ in the class C such that the image of L under the evaluation map is $G \setminus \{1_G\}$. We provide some examples of epiC groups and prove that the property of being epiC is not dependent on the generating set chosen. We also prove that epiC groups are closed under passage to finite index overgroups, taking extensions, and taking graph products of finitely many groups. Furthermore, we prove that epiRegular groups are closed under passage to finite index subgroups. Finally, we provide a characterization of the property of having solvable word problem within the framework of epiC groups., Comment: 21 pages, 0 figures

Quantum error correction is essential for bridging the gap between the error rates of physical devices and the extremely low logical error rates required for quantum algorithms. Recent error-correction demonstrations on superconducting processors have focused primarily on the surface code, which offers a high error threshold but poses limitations for logical operations. In contrast, the color code enables much more efficient logic, although it requires more complex stabilizer measurements and decoding techniques. Measuring these stabilizers in planar architectures such as superconducting qubits is challenging, and so far, realizations of color codes have not addressed performance scaling with code size on any platform. Here, we present a comprehensive demonstration of the color code on a superconducting processor, achieving logical error suppression and performing logical operations. Scaling the code distance from three to five suppresses logical errors by a factor of $\Lambda_{3/5}$ = 1.56(4). Simulations indicate this performance is below the threshold of the color code, and furthermore that the color code may be more efficient than the surface code with modest device improvements. Using logical randomized benchmarking, we find that transversal Clifford gates add an error of only 0.0027(3), which is substantially less than the error of an idling error correction cycle. We inject magic states, a key resource for universal computation, achieving fidelities exceeding 99% with post-selection (retaining about 75% of the data). Finally, we successfully teleport logical states between distance-three color codes using lattice surgery, with teleported state fidelities between 86.5(1)% and 90.7(1)%. This work establishes the color code as a compelling research direction to realize fault-tolerant quantum computation on superconducting processors in the near future.

This paper addresses the problem of detecting points on or near the boundary of a dataset sampled, potentially with noise, from a compact manifold with boundary. We extend recent advances in doubly stochastic scaling of the Gaussian heat kernel via Sinkhorn iterations to this setting. Our main contributions are: (a) deriving a characterization of the scaling factors for manifolds with boundary, (b) developing a boundary direction estimator, aimed at identifying boundary points, based on doubly stochastic kernel and local principal component analysis, and (c) demonstrating through simulations that the resulting estimates of the boundary points outperform the standard Gaussian kernel-based approach, particularly under noisy conditions.

Social impact evaluations are emerging as a useful tool to understand, document, and evaluate the societal impacts of generative AI. In this provocation, we begin to think carefully about the types of experts and expertise that are needed to conduct robust social impact evaluations of generative AI. We suggest that doing so will require thoughtfully eliciting and integrating insights from a range of "domain experts" and "experiential experts," and close with five open questions.

Experience goods such as sporting and artistic events, songs, videos, news stories, podcasts, and television series, are often packaged and consumed in bundles. Many such bundles are ordered in the sense that the individual items are consumed sequentially, one at a time. We examine if an individual's decision to consume the next item in an ordered bundle can be predicted based on his/her consumption pattern for the preceding items. We evaluate several predictive models, including two custom Transformers using decoder-only and encoder-decoder architectures, fine-tuned GPT-3, a custom LSTM model, a reinforcement learning model, two Markov models, and a zero-order model. Using data from Spotify, we find that the custom Transformer with a decoder-only architecture provides the most accurate predictions, both for individual choices and aggregate demand. This model captures a general form of state dependence. Analysis of Transformer attention weights suggests that the consumption of the next item in a bundle is based on approximately equal weighting of all preceding choices. Our results indicate that the Transformer can assist in queuing the next item that an individual is likely to consume from an ordered bundle, predicting the demand for individual items, and personalizing promotions to increase demand.

Wearable sensors have become ubiquitous thanks to a variety of health tracking features. The resulting continuous and longitudinal measurements from everyday life generate large volumes of data; however, making sense of these observations for scientific and actionable insights is non-trivial. Inspired by the empirical success of generative modeling, where large neural networks learn powerful representations from vast amounts of text, image, video, or audio data, we investigate the scaling properties of sensor foundation models across compute, data, and model size. Using a dataset of up to 40 million hours of in-situ heart rate, heart rate variability, electrodermal activity, accelerometer, skin temperature, and altimeter per-minute data from over 165,000 people, we create LSM, a multimodal foundation model built on the largest wearable-signals dataset with the most extensive range of sensor modalities to date. Our results establish the scaling laws of LSM for tasks such as imputation, interpolation and extrapolation, both across time and sensor modalities. Moreover, we highlight how LSM enables sample-efficient downstream learning for tasks like exercise and activity recognition.

Passively collected "big" data sources are increasingly used to inform critical development policy decisions in low- and middle-income countries. While prior work highlights how such approaches may reveal sensitive information, enable surveillance, and centralize power, less is known about the corresponding privacy concerns, hopes, and fears of the people directly impacted by these policies -- people sometimes referred to as experiential experts. To understand the perspectives of experiential experts, we conducted semi-structured interviews with people living in rural villages in Togo shortly after an entirely digital cash transfer program was launched that used machine learning and mobile phone metadata to determine program eligibility. This paper documents participants' privacy concerns surrounding the introduction of big data approaches in development policy. We find that the privacy concerns of our experiential experts differ from those raised by privacy and development domain experts. To facilitate a more robust and constructive account of privacy, we discuss implications for policies and designs that take seriously the privacy concerns raised by both experiential experts and domain experts.

Computational design of protein-binding proteins is a fundamental capability with broad utility in biomedical research and biotechnology. Recent methods have made strides against some target proteins, but on-demand creation of high-affinity binders without multiple rounds of experimental testing remains an unsolved challenge. This technical report introduces AlphaProteo, a family of machine learning models for protein design, and details its performance on the de novo binder design problem. With AlphaProteo, we achieve 3- to 300-fold better binding affinities and higher experimental success rates than the best existing methods on seven target proteins. Our results suggest that AlphaProteo can generate binders "ready-to-use" for many research applications using only one round of medium-throughput screening and no further optimization., Comment: 45 pages, 17 figures

The Internet of Things (IoT) is a network of billions of interconnected, primarily low-end embedded devices. Despite large-scale deployment, studies have highlighted critical security concerns in IoT networks, many of which stem from firmware-related issues. Furthermore, IoT swarms have become more prevalent in industries, smart homes, and agricultural applications, among others. Malicious activity on one node in a swarm can propagate to larger network sections. Although several Remote Attestation (RA) techniques have been proposed, they are limited by their latency, availability, complexity, hardware assumptions, and uncertain access to firmware copies under Intellectual Property (IP) rights. We present Swarm-Net, a novel swarm attestation technique that exploits the inherent, interconnected, graph-like structure of IoT networks along with the runtime information stored in the Static Random Access Memory (SRAM) using Graph Neural Networks (GNN) to detect malicious firmware and its downstream effects. We also present the first datasets on SRAM-based swarm attestation encompassing different types of firmware and edge relationships. In addition, a secure swarm attestation protocol is presented. Swarm-Net is not only computationally lightweight but also does not require a copy of the firmware. It achieves a 99.96% attestation rate on authentic firmware, 100% detection rate on anomalous firmware, and 99% detection rate on propagated anomalies, at a communication overhead and inference latency of ~1 second and ~10^{-5} seconds (on a laptop CPU), respectively. In addition to the collected datasets, Swarm-Net's effectiveness is evaluated on simulated trace replay, random trace perturbation, and dropped attestation responses, showing robustness against such threats. Lastly, we compare Swarm-Net with past works and present a security analysis., Comment: Under Review

Quantum error correction provides a path to reach practical quantum computing by combining multiple physical qubits into a logical qubit, where the logical error rate is suppressed exponentially as more qubits are added. However, this exponential suppression only occurs if the physical error rate is below a critical threshold. In this work, we present two surface code memories operating below this threshold: a distance-7 code and a distance-5 code integrated with a real-time decoder. The logical error rate of our larger quantum memory is suppressed by a factor of $\Lambda$ = 2.14 $\pm$ 0.02 when increasing the code distance by two, culminating in a 101-qubit distance-7 code with 0.143% $\pm$ 0.003% error per cycle of error correction. This logical memory is also beyond break-even, exceeding its best physical qubit's lifetime by a factor of 2.4 $\pm$ 0.3. We maintain below-threshold performance when decoding in real time, achieving an average decoder latency of 63 $\mu$s at distance-5 up to a million cycles, with a cycle time of 1.1 $\mu$s. To probe the limits of our error-correction performance, we run repetition codes up to distance-29 and find that logical performance is limited by rare correlated error events occurring approximately once every hour, or 3 $\times$ 10$^9$ cycles. Our results present device performance that, if scaled, could realize the operational requirements of large scale fault-tolerant quantum algorithms., Comment: 10 pages, 4 figures, Supplementary Information

With recent rapid growth in online shopping, AI-powered Engagement Surfaces (ES) have become ubiquitous across retail services. These engagement surfaces perform an increasing range of functions, including recommending new products for purchase, reminding customers of their orders and providing delivery notifications. Understanding the causal effect of engagement surfaces on value driven for customers and businesses remains an open scientific question. In this paper, we develop a dynamic causal model at scale to disentangle value attributable to an ES, and to assess its effectiveness. We demonstrate the application of this model to inform business decision-making by understanding returns on investment in the ES, and identifying product lines and features where the ES adds the most value., Comment: 10 pages, 5 figures. Accepted at Applied Data Science track of KDD 2024, Barcelona, Spain

We study groups whose co-word problems are ET0L languages, which we call coET0L groups, using an automaton based model due to van Leeuwen, and recently studied by Bishop and Elder. In particular we prove a number of closure results for the class of groups with co-word problems in a subclass of `special' ET0L languages; that class of groups contains all groups that we know at the time of writing to be co-ET0L, including all groups that were proved by Holt and R\"over to be stack groups, and hence co-indexed. It includes virtually free groups, bounded automata groups, and the Higman-Thompson groups, together with groups constructed from those using finitely generated subgroups, finite extension, free and direct products, and by taking the restricted standard wreath product of a co-\E group by a finitely generated virtually free top group.

Advanced AI assistants combine frontier LLMs and tool access to autonomously perform complex tasks on behalf of users. While the helpfulness of such assistants can increase dramatically with access to user information including emails and documents, this raises privacy concerns about assistants sharing inappropriate information with third parties without user supervision. To steer information-sharing assistants to behave in accordance with privacy expectations, we propose to operationalize contextual integrity (CI), a framework that equates privacy with the appropriate flow of information in a given context. In particular, we design and evaluate a number of strategies to steer assistants' information-sharing actions to be CI compliant. Our evaluation is based on a novel form filling benchmark composed of human annotations of common webform applications, and it reveals that prompting frontier LLMs to perform CI-based reasoning yields strong results.

Modern artificial intelligence (AI) systems are powered by foundation models. This paper presents a new set of foundation models, called Llama 3. It is a herd of language models that natively support multilinguality, coding, reasoning, and tool usage. Our largest model is a dense Transformer with 405B parameters and a context window of up to 128K tokens. This paper presents an extensive empirical evaluation of Llama 3. We find that Llama 3 delivers comparable quality to leading language models such as GPT-4 on a plethora of tasks. We publicly release Llama 3, including pre-trained and post-trained versions of the 405B parameter language model and our Llama Guard 3 model for input and output safety. The paper also presents the results of experiments in which we integrate image, video, and speech capabilities into Llama 3 via a compositional approach. We observe this approach performs competitively with the state-of-the-art on image, video, and speech recognition tasks. The resulting models are not yet being broadly released as they are still under development.

The Pockels effect is essential for controlling optical signals at the highest speeds. We present the first measurements of the Pockels coefficients and permittivity in lithium niobate (LN) and barium titanate (BTO) over a continuous frequency range from 100 MHz to 330 GHz. These properties are constant across this frequency range in LN but have a significant frequency dependence in BTO. Still, our measurements show that BTO ($\varepsilon$ = 1136, $r_{42}$ = 481 pm/V, $r_{33}$ = 125 pm/V at 100 MHz, $\varepsilon$ = 453, $r_{42}$ = 191 pm/V, $r_{33}$ = 60 pm/V at 330 GHz) has remarkably large electro-optic properties compared to LN ($\varepsilon$ = 27, $r_{42}$ = 15 pm/V, $r_{33}$ = 27 pm/V). Furthermore, we show how BTO devices can be designed with a flat electro-optic frequency response despite the Pockels coefficient dispersion. Finally, we expound our method for broadband characterization of these vital electro-optic properties, utilizing specialized integrated electro-optic phase shifters. Altogether, this work is foundational to designing high-speed BTO devices and to developing new electro-optic materials., Comment: Main Text: 19 pages, 5 figures. Supplementary Information: 18 pages, 10 figures

Epilepsy is the most common, chronic, neurological disease worldwide and is typically accompanied by reoccurring seizures. Neuro implants can be used for effective treatment by suppressing an upcoming seizure upon detection. Due to the restricted size and limited battery lifetime of those medical devices, the employed approach also needs to be limited in size and have low energy requirements. We present an energy-efficient seizure detection approach involving a TC-ResNet and time-series analysis which is suitable for low-power edge devices. The presented approach allows for accurate seizure detection without preceding feature extraction while considering the stringent hardware requirements of neural implants. The approach is validated using the CHB-MIT Scalp EEG Database with a 32-bit floating point model and a hardware suitable 4-bit fixed point model. The presented method achieves an accuracy of 95.28%, a sensitivity of 92.34% and an AUC score of 0.9384 on this dataset with 4-bit fixed point representation. Furthermore, the power consumption of the model is measured with the low-power AI accelerator UltraTrail, which only requires 495 nW on average. Due to this low-power consumption this classification approach is suitable for real-time seizure detection on low-power wearable devices such as neural implants., Comment: Accepted at IJCNN 2024 (Preprint)

We present TORAX, a new, open-source, differentiable tokamak core transport simulator implemented in Python using the JAX framework. TORAX solves the coupled equations for ion heat transport, electron heat transport, particle transport, and current diffusion, incorporating modular physics-based and ML models. JAX's just-in-time compilation ensures fast runtimes, while its automatic differentiation capability enables gradient-based optimization workflows and simplifies the use of Jacobian-based PDE solvers. Coupling to ML-surrogates of physics models is greatly facilitated by JAX's intrinsic support for neural network development and inference. TORAX is verified against the established RAPTOR code, demonstrating agreement in simulated plasma profiles. TORAX provides a powerful and versatile tool for accelerating research in tokamak scenario modeling, pulse design, and control., Comment: 16 pages, 7 figures

The rapid progress of large language models (LLMs) has seen them excel and frequently surpass human performance on standard benchmarks. This has enabled many downstream applications, such as LLM agents, to rely on their sophisticated reasoning to navigate complex task requirements. However, LLMs are known to unexpectedly falter in simple tasks and under seemingly straightforward circumstances - underscoring the need for better and more diverse evaluation setups to measure their true capabilities. To this end, we choose to study compositional and conditional reasoning, two cornerstones of human cognition, and introduce GroundCocoa - a lexically diverse benchmark connecting these reasoning skills to the real-world problem of flight booking. Our task involves aligning detailed user preferences with available flight options presented in a multiple-choice format. Results indicate a significant disparity in performance among current state-of-the-art LLMs with even the best performing model, GPT-4 Turbo, not exceeding 67% accuracy despite advanced prompting techniques., Comment: 18 pages, 17 figures, 3 tables. Paper under review

In contemporary machine learning approaches to bilingual lexicon induction (BLI), a model learns a mapping between the embedding spaces of a language pair. Recently, retrieve-and-rank approach to BLI has achieved state of the art results on the task. However, the problem remains challenging in low-resource settings, due to the paucity of data. The task is complicated by factors such as lexical variation across languages. We argue that the incorporation of additional lexical information into the recent retrieve-and-rank approach should improve lexicon induction. We demonstrate the efficacy of our proposed approach on XLING, improving over the previous state of the art by an average of 2\% across all language pairs., Comment: 8 pages, 4 figures. Paper accepted at NAACL Findings 2024

Event cameras, also known as dynamic vision sensors, are an emerging modality for measuring fast dynamics asynchronously. Event cameras capture changes of log-intensity over time as a stream of 'events' and generally cannot measure intensity itself; hence, they are only used for imaging dynamic scenes. However, fluctuations due to random photon arrival inevitably trigger noise events, even for static scenes. While previous efforts have been focused on filtering out these undesirable noise events to improve signal quality, we find that, in the photon-noise regime, these noise events are correlated with the static scene intensity. We analyze the noise event generation and model its relationship to illuminance. Based on this understanding, we propose a method, called Noise2Image, to leverage the illuminance-dependent noise characteristics to recover the static parts of a scene, which are otherwise invisible to event cameras. We experimentally collect a dataset of noise events on static scenes to train and validate Noise2Image. Our results provide a novel approach for capturing static scenes in event cameras, solely from noise events, without additional hardware.