Search

In this paper, we show that every pair of large even integers satisfying certain necessary conditions can be expressed as a pair of one prime, one prime square, two prime cubes and 56 powers of 2.

Many molecular alterations serve as clinically prognostic or therapy-predictive biomarkers, typically detected using single or multi-gene molecular assays. However, these assays are expensive, tissue destructive and often take weeks to complete. Using AI on routine H&E WSIs offers a fast and economical approach to screen for multiple molecular biomarkers. We present a high-throughput AI-based system leveraging Virchow2, a foundation model pre-trained on 3 million slides, to interrogate genomic features previously determined by an next-generation sequencing (NGS) assay, using 47,960 scanned hematoxylin and eosin (H&E) whole slide images (WSIs) from 38,984 cancer patients. Unlike traditional methods that train individual models for each biomarker or cancer type, our system employs a unified model to simultaneously predict a wide range of clinically relevant molecular biomarkers across cancer types. By training the network to replicate the MSK-IMPACT targeted biomarker panel of 505 genes, it identified 80 high performing biomarkers with a mean AU-ROC of 0.89 in 15 most common cancer types. In addition, 40 biomarkers demonstrated strong associations with specific cancer histologic subtypes. Furthermore, 58 biomarkers were associated with targets frequently assayed clinically for therapy selection and response prediction. The model can also predict the activity of five canonical signaling pathways, identify defects in DNA repair mechanisms, and predict genomic instability measured by tumor mutation burden, microsatellite instability (MSI), and chromosomal instability (CIN). The proposed model can offer potential to guide therapy selection, improve treatment efficacy, accelerate patient screening for clinical trials and provoke the interrogation of new therapeutic targets.

Foundation models are rapidly being developed for computational pathology applications. However, it remains an open question which factors are most important for downstream performance with data scale and diversity, model size, and training algorithm all playing a role. In this work, we propose algorithmic modifications, tailored for pathology, and we present the result of scaling both data and model size, surpassing previous studies in both dimensions. We introduce two new models: Virchow2, a 632 million parameter vision transformer, and Virchow2G, a 1.9 billion parameter vision transformer, each trained with 3.1 million histopathology whole slide images, with diverse tissues, originating institutions, and stains. We achieve state of the art performance on 12 tile-level tasks, as compared to the top performing competing models. Our results suggest that data diversity and domain-specific methods can outperform models that only scale in the number of parameters, but, on average, performance benefits from the combination of domain-specific methods, data scale, and model scale.

Foundation models in computational pathology promise to unlock the development of new clinical decision support systems and models for precision medicine. However, there is a mismatch between most clinical analysis, which is defined at the level of one or more whole slide images, and foundation models to date, which process the thousands of image tiles contained in a whole slide image separately. The requirement to train a network to aggregate information across a large number of tiles in multiple whole slide images limits these models' impact. In this work, we present a slide-level foundation model for H&E-stained histopathology, PRISM, that builds on Virchow tile embeddings and leverages clinical report text for pre-training. Using the tile embeddings, PRISM produces slide-level embeddings with the ability to generate clinical reports, resulting in several modes of use. Using text prompts, PRISM achieves zero-shot cancer detection and sub-typing performance approaching and surpassing that of a supervised aggregator model. Using the slide embeddings with linear classifiers, PRISM surpasses supervised aggregator models. Furthermore, we demonstrate that fine-tuning of the PRISM slide encoder yields label-efficient training for biomarker prediction, a task that typically suffers from low availability of training data; an aggregator initialized with PRISM and trained on as little as 10% of the training data can outperform a supervised baseline that uses all of the data.

Self-supervised learning (SSL) has emerged as a key technique for training networks that can generalize well to diverse tasks without task-specific supervision. This property makes SSL desirable for computational pathology, the study of digitized images of tissues, as there are many target applications and often limited labeled training samples. However, SSL algorithms and models have been primarily developed in the field of natural images and whether their performance can be improved by adaptation to particular domains remains an open question. In this work, we present an investigation of modifications to SSL for pathology data, specifically focusing on the DINOv2 algorithm. We propose alternative augmentations, regularization functions, and position encodings motivated by the characteristics of pathology images. We evaluate the impact of these changes on several benchmarks to demonstrate the value of tailored approaches., Comment: Presented at DCA in MI Workshop, CVPR 2024

Normal-form games with two players, each with two strategies, are the most studied class of games. These so-called 2x2 games are used to model a variety of strategic interactions. They appear in game theory, economics, and artificial intelligence research. However, there lacks tools for describing and visualizing such games. This work introduces a LaTeX package for visualizing 2x2 games. This work has two goals: first, to provide high-quality tools and vector graphic visualizations, suitable for scientific publications. And second, to help promote standardization of names and representations of 2x2 games. The LaTeX package, twoxtwogame, is maintained on GitHub and mirrored on CTAN, and is available under a permissive Apache 2 license.

Normal-form games (NFGs) are the fundamental model of strategic interaction. We study their representation using neural networks. We describe the inherent equivariance of NFGs -- any permutation of strategies describes an equivalent game -- as well as the challenges this poses for representation learning. We then propose the NfgTransformer architecture that leverages this equivariance, leading to state-of-the-art performance in a range of game-theoretic tasks including equilibrium-solving, deviation gain estimation and ranking, with a common approach to NFG representation. We show that the resulting model is interpretable and versatile, paving the way towards deep learning systems capable of game-theoretic reasoning when interacting with humans and with each other., Comment: Published at ICLR 2024. Open-sourced at https://github.com/google-deepmind/nfg_transformer

Game theory is the study of mathematical models of strategic interactions among rational agents. Language is a key medium of interaction for humans, though it has historically proven difficult to model dialogue and its strategic motivations mathematically. A suitable model of the players, strategies, and payoffs associated with linguistic interactions (i.e., a binding to the conventional symbolic logic of game theory) would enable existing game-theoretic algorithms to provide strategic solutions in the space of language. In other words, a binding could provide a route to computing stable, rational conversational strategies in dialogue. Large language models (LLMs) have arguably reached a point where their generative capabilities can enable realistic, human-like simulations of natural dialogue. By prompting them in various ways, we can steer their responses towards different output utterances. Leveraging the expressivity of natural language, LLMs can also help us quickly generate new dialogue scenarios, which are grounded in real world applications. In this work, we present one possible binding from dialogue to game theory as well as generalizations of existing equilibrium finding algorithms to this setting. In addition, by exploiting LLMs generation capabilities along with our proposed binding, we can synthesize a large repository of formally-defined games in which one can study and test game-theoretic solution concepts. We also demonstrate how one can combine LLM-driven game generation, game-theoretic solvers, and imitation learning to construct a process for improving the strategic capabilities of LLMs., Comment: 32 pages, 8 figures, code available @ https://github.com/google-deepmind/open_spiel/blob/master/open_spiel/python/games/chat_game.py

We study computationally efficient methods for finding equilibria in n-player general-sum games, specifically ones that afford complex visuomotor skills. We show how existing methods would struggle in this setting, either computationally or in theory. We then introduce NeuPL-JPSRO, a neural population learning algorithm that benefits from transfer learning of skills and converges to a Coarse Correlated Equilibrium (CCE) of the game. We show empirical convergence in a suite of OpenSpiel games, validated rigorously by exact game solvers. We then deploy NeuPL-JPSRO to complex domains, where our approach enables adaptive coordination in a MuJoCo control domain and skill transfer in capture-the-flag. Our work shows that equilibrium convergent population learning can be implemented at scale and in generality, paving the way towards solving real-world games between heterogeneous players with mixed motives.

Mindfulness-based therapies have been shown to be effective in improving mental health, and technology-based methods have the potential to expand the accessibility of these therapies. To enable real-time personalized content generation for mindfulness practice in these methods, high-quality computer-synthesized text-to-speech (TTS) voices are needed to provide verbal guidance and respond to user performance and preferences. However, the user-perceived quality of state-of-the-art TTS voices has not yet been evaluated for administering mindfulness meditation, which requires emotional expressiveness. In addition, work has not yet been done to study the effect of physical embodiment and personalization on the user-perceived quality of TTS voices for mindfulness. To that end, we designed a two-phase human subject study. In Phase 1, an online Mechanical Turk between-subject study (N=471) evaluated 3 (feminine, masculine, child-like) state-of-the-art TTS voices with 2 (feminine, masculine) human therapists' voices in 3 different physical embodiment settings (no agent, conversational agent, socially assistive robot) with remote participants. Building on findings from Phase 1, in Phase 2, an in-person within-subject study (N=94), we used a novel framework we developed for personalizing TTS voices based on user preferences, and evaluated user-perceived quality compared to best-rated non-personalized voices from Phase 1. We found that the best-rated human voice was perceived better than all TTS voices; the emotional expressiveness and naturalness of TTS voices were poorly rated, while users were satisfied with the clarity of TTS voices. Surprisingly, by allowing users to fine-tune TTS voice features, the user-personalized TTS voices could perform almost as well as human voices, suggesting user personalization could be a simple and very effective tool to improve user-perceived quality of TTS voice., Comment: Published in Proceedings of the 2023 ACM/IEEE International Conference on Human-Robot Interaction, pp. 516-524. 2023

As artificial intelligence (AI) is playing an increasingly important role in our society and global economy, AI education and literacy have become necessary components in college and K-12 education to prepare students for an AI-powered society. However, current AI curricula have not yet been made accessible and engaging enough for students and schools from all socio-economic backgrounds with different educational goals. In this work, we developed an open-source learning module for college and high school students, which allows students to build their own robot companion from the ground up. This open platform can be used to provide hands-on experience and introductory knowledge about various aspects of AI, including robotics, machine learning (ML), software engineering, and mechanical engineering. Because of the social and personal nature of a socially assistive robot companion, this module also puts a special emphasis on human-centered AI, enabling students to develop a better understanding of human-AI interaction and AI ethics through hands-on learning activities. With open-source documentation, assembling manuals and affordable materials, students from different socio-economic backgrounds can personalize their learning experience based on their individual educational goals. To evaluate the student-perceived quality of our module, we conducted a usability testing workshop with 15 college students recruited from a minority-serving institution. Our results indicate that our AI module is effective, easy-to-follow, and engaging, and it increases student interest in studying AI/ML and robotics in the future. We hope that this work will contribute toward accessible and engaging AI education in human-AI interaction for college and high school students., Comment: Accepted to the Proceedings of the AAAI Conference on Artificial Intelligence (2024)

Embedding Human and Articulated Object Interaction (HAOI) in 3D is an important direction for a deeper human activity understanding. Different from previous works that use parametric and CAD models to represent humans and objects, in this work, we propose a novel 3D geometric primitive-based language to encode both humans and objects. Given our new paradigm, humans and objects are all compositions of primitives instead of heterogeneous entities. Thus, mutual information learning may be achieved between the limited 3D data of humans and different object categories. Moreover, considering the simplicity of the expression and the richness of the information it contains, we choose the superquadric as the primitive representation. To explore an effective embedding of HAOI for the machine, we build a new benchmark on 3D HAOI consisting of primitives together with their images and propose a task requiring machines to recover 3D HAOI using primitives from images. Moreover, we propose a baseline of single-view 3D reconstruction on HAOI. We believe this primitive-based 3D HAOI representation would pave the way for 3D HAOI studies. Our code and data are available at https://mvig-rhos.com/p3haoi., Comment: AAAI2024

Name-based access control (NAC) over NDN provides fine-grained data confidentiality and access control by encrypting and signing data at the time of data production. NAC utilizes specially crafted naming conventions to define and enforce access control policies. NAC-ABE, an extension to NAC, uses an attribute-based encryption (ABE) scheme to support access control with improved scalability and flexibility. However, existing NAC-ABE libraries are based on ciphertext-policy ABE (CP-ABE), which requires knowledge of the access policy when encrypting data packets. In some applications, including mHealth, the data access policy is unknown at the time of data generation, while data attributes and properties are known. In this paper, we present an extension to the existing NDN-ABE library which can be used by mHealth and other applications to enforce fine-granularity access control in data sharing. We also discuss the challenges we encountered during the application deployment, and remaining open issues together with potential solution directions.

In 2021, a catalog of 536 fast radio bursts (FRBs) detected with the Canadian Hydrogen Intensity Mapping Experiment (CHIME) radio telescope was released by the CHIME/FRB Collaboration. This large collection of bursts, observed with a single instrument and uniform selection effects, has advanced our understanding of the FRB population. Here we update the results for 140 of these FRBs for which channelized raw voltage ('baseband') data are available. With the voltages measured by the telescope's antennas, it is possible to maximize the telescope sensitivity in any direction within the primary beam, an operation called 'beamforming'. This allows us to increase the signal-to-noise ratio (S/N) of the bursts and to localize them to sub-arcminute precision. The improved localization is also used to correct the beam response of the instrument and to measure fluxes and fluences with a ~10% uncertainty. Additionally, the time resolution is increased by three orders of magnitude relative to that in the first CHIME/FRB catalog, and, applying coherent dedispersion, burst morphologies can be studied in detail. Polarization information is also available for the full sample of 140 FRBs, providing an unprecedented dataset to study the polarization properties of the population. We release the baseband data beamformed to the most probable position of each FRB. These data are analyzed in detail in a series of accompanying papers.

Cell segmentation for multi-modal microscopy images remains a challenge due to the complex textures, patterns, and cell shapes in these images. To tackle the problem, we first develop an automatic cell classification pipeline to label the microscopy images based on their low-level image characteristics, and then train a classification model based on the category labels. Afterward, we train a separate segmentation model for each category using the images in the corresponding category. Besides, we further deploy two types of segmentation models to segment cells with roundish and irregular shapes respectively. Moreover, an efficient and powerful backbone model is utilized to enhance the efficiency of our segmentation model. Evaluated on the Tuning Set of NeurIPS 2022 Cell Segmentation Challenge, our method achieves an F1-score of 0.8795 and the running time for all cases is within the time tolerance., Comment: The second place in NeurIPS 2022 cell segmentation challenge (https://neurips22-cellseg.grand-challenge.org/), released code: https://github.com/lhaof/CellSeg

Nuclei segmentation is a fundamental but challenging task in the quantitative analysis of histopathology images. Although fully-supervised deep learning-based methods have made significant progress, a large number of labeled images are required to achieve great segmentation performance. Considering that manually labeling all nuclei instances for a dataset is inefficient, obtaining a large-scale human-annotated dataset is time-consuming and labor-intensive. Therefore, augmenting a dataset with only a few labeled images to improve the segmentation performance is of significant research and application value. In this paper, we introduce the first diffusion-based augmentation method for nuclei segmentation. The idea is to synthesize a large number of labeled images to facilitate training the segmentation model. To achieve this, we propose a two-step strategy. In the first step, we train an unconditional diffusion model to synthesize the Nuclei Structure that is defined as the representation of pixel-level semantic and distance transform. Each synthetic nuclei structure will serve as a constraint on histopathology image synthesis and is further post-processed to be an instance map. In the second step, we train a conditioned diffusion model to synthesize histopathology images based on nuclei structures. The synthetic histopathology images paired with synthetic instance maps will be added to the real dataset for training the segmentation model. The experimental results show that by augmenting 10% labeled real dataset with synthetic samples, one can achieve comparable segmentation results with the fully-supervised baseline. The code is released in: https://github.com/lhaof/Nudiff, Comment: MICCAI 2023, released code: https://github.com/lhaof/Nudiff

The use of artificial intelligence to enable precision medicine and decision support systems through the analysis of pathology images has the potential to revolutionize the diagnosis and treatment of cancer. Such applications will depend on models' abilities to capture the diverse patterns observed in pathology images. To address this challenge, we present Virchow, a foundation model for computational pathology. Using self-supervised learning empowered by the DINOv2 algorithm, Virchow is a vision transformer model with 632 million parameters trained on 1.5 million hematoxylin and eosin stained whole slide images from diverse tissue and specimen types, which is orders of magnitude more data than previous works. The Virchow model enables the development of a pan-cancer detection system with 0.949 overall specimen-level AUC across 17 different cancer types, while also achieving 0.937 AUC on 7 rare cancer types. The Virchow model sets the state-of-the-art on the internal and external image tile level benchmarks and slide level biomarker prediction tasks. The gains in performance highlight the importance of training on massive pathology image datasets, suggesting scaling up the data and network architecture can improve the accuracy for many high-impact computational pathology applications where limited amounts of training data are available.

We describe a new parameterized family of symmetric error-correcting codes with low-density parity-check matrices (LDPC). Our codes can be described in two seemingly different ways. First, in relation to Reed-Muller codes: our codes are functions on a subset of $\mathbb{F}^n$ whose restrictions to a prescribed set of affine lines has low degree. Alternatively, they are Tanner codes on high dimensional expanders, where the coordinates of the codeword correspond to triangles of a $2$-dimensional expander, such that around every edge the local view forms a Reed-Solomon codeword. For some range of parameters our codes are provably locally testable, and their dimension is some fixed power of the block length. For another range of parameters our codes have distance and dimension that are both linear in the block length, but we do not know if they are locally testable. The codes also have the multiplication property: the coordinate-wise product of two codewords is a codeword in a related code. The definition of the codes relies on the construction of a specific family of simplicial complexes which is a slight variant on the coset complexes of Kaufman and Oppenheim. We show a novel way to embed the triangles of these complexes into $\mathbb{F}^n$, with the property that links of edges embed as affine lines in $\mathbb{F}^n$. We rely on this embedding to lower bound the rate of these codes in a way that avoids constraint-counting and thereby achieves non-trivial rate even when the local codes themselves have arbitrarily small rate, and in particular below $1/2$.

The study of massive neutrinos and their interactions is a critical aspect of contemporary cosmology. Recent advances in parallel computation and high-performance computing provide new opportunities for accurately constraining Large-Scale Structures (LSS). In this paper, we introduce the TianNu cosmological N-body simulation during the co-evolution of massive neutrino and cold dark matter components via the CUBEP$^3$M code running on the supercomputer Tianhe-2 and TianNu's connected works. We start by analyzing $2.537\times10^7$ dark halos from the scientific data of TianNu simulation, and compare their angular momentum with the matched halos from neutrino-free TianZero, revealing a dependence of angular momentum modulus on neutrino injection at scales below 50 Mpc and around 10 Mpc., Comment: 11 pages(without ref.), 10 figures, 2 tables. The review of related works about the TianNu simulation

Making image retrieval methods practical for real-world search applications requires significant progress in dataset scales, entity comprehension, and multimodal information fusion. In this work, we introduce \textbf{E}ntity-\textbf{D}riven \textbf{I}mage \textbf{S}earch (EDIS), a challenging dataset for cross-modal image search in the news domain. EDIS consists of 1 million web images from actual search engine results and curated datasets, with each image paired with a textual description. Unlike datasets that assume a small set of single-modality candidates, EDIS reflects real-world web image search scenarios by including a million multimodal image-text pairs as candidates. EDIS encourages the development of retrieval models that simultaneously address cross-modal information fusion and matching. To achieve accurate ranking results, a model must: 1) understand named entities and events from text queries, 2) ground entities onto images or text descriptions, and 3) effectively fuse textual and visual representations. Our experimental results show that EDIS challenges state-of-the-art methods with dense entities and a large-scale candidate set. The ablation study also proves that fusing textual features with visual features is critical in improving retrieval results., Comment: EMNLP 2023 camera ready version

This article reports a sample of 1830 strong [O III] {\lambda}5007 emission-line compact galaxies discovered with the LAMOST spectroscopic survey and the photometric catalog of SDSS. We newly identify 402 spectra of 346 strong [O III]{\lambda}5007 emission-line compact galaxies by finding compact isolated point sources. Combined with the samples in our previous work (Liu et al. 2022), this returns a sample of 1830 unique strong [O III]{\lambda}5007 emission-line compact galaxies with 2033 spectra of z <= 0.53. For the sources with 2{\sigma} [OIII]{\lambda}4363 detections, we calculate the gas-phase metallicity with the direct-Te method, and verify that the strong-line metallicity diagnostics calibrated with the direct-Te method also applies to this sample. The strong [O III]{\lambda}5007 emission-line compact galaxies fall below several Te-calibrated mass-metallicity relations. The N/O measurements of the strong [O iii]{\lambda}5007 emission-line compact galaxies mainly locate at a plateau at low metallicity, indicating the product of primary nucleosynthesis. The Ne3O2 and O32 relation follows a tight linear relation with no redshift evolution. The Ne3O2 anti-correlates with the stellar mass, and at fixed stellar mass the Ne3O2 increase with the redshift. Eight sources with asymmetric [O III]{\lambda}5007 emission-line profiles have been identified, however with no [O III]{\lambda}4363 detection, which proves the rich metal content and complex ionized gas kinematics within the galaxies. Higher-resolution spectroscopy will be necessary to identify the ionized gas components in detail., Comment: 20 pages, 13 pictures, accepted by ApJS

Green Pea galaxies are compact galaxies with high star formation rates. However, limited samples of Green Pea galaxies have HI 21 cm measurements. Whether the HI gas fraction f_{HI} = M_{HI}/M_{*} of Green Pea galaxies follows the existing scaling relations between the f_{HI} and NUV-r color or linear combinations of color and other physical quantities needs checking. Using archival data of HI 21cm observations, we investigate the scaling relation of the NUV-r color with the M_{HI}/M_{*} of 38 Green Pea galaxies, including 17 detections and 21 non-detections. The HI to stellar mass ratios (f_{HI}) of Green Pea galaxies deviate from the polynomial form, where a higher HI gas fraction is predicted given the current NUV-r color, even with the emission lines removed. The blue sources (NUV-r<1) from the comparison sample (ALFALFA-SDSS) follow a similar trend. The HI gas fraction scaling relations with linear combination forms of -0.34(NUV-r) - 0.64 log(mu_{*,z}) + 5.94 and -0.77 log mu_{*,i} + 0.26 log SFR/M_{*}+8.53, better predict the HI gas fraction of the Green Pea galaxies. In order to obtain accurate linear combined forms, higher-resolution photometry from space-based telescopes is needed., Comment: 15 pages, 7 figures, to be published in RAA

We report the discovery of a 100 kpc HI tail in the merging galaxy pair NGC 4490/85 detected by the Five-Hundred-meter Aperture Spherical radio Telescope (FAST). The tidal tails extended in both the south and north directions, and they are much longer than that reported previously based on the VLA interferometric maps. The NGC 4490/85 is surrounded by a large gas envelope, and a starburst low metallicity dwarf galaxy MAPS 1231+42 is found to be connected with the gas envelope, indicating that galaxy interaction trigged the intense star formation in it. Based on the fact that the metallicity in MAPS 1231+42 is one order of magnitude lower than that in the two disks of NGC 4490 and NGC 4485, we speculate that the gas near this galaxy should be primordial and could be due to gas inflow from the circum-galactic medium (CGM). We also found a collimated gas component pointing at a nearby dwarf galaxy KK 149, suggesting that this galaxy might also be interacting with the NGC 4490 pair. We discuss the possible origin of the long tidal tails and the extended gas envelope in this merging system based on the new data from FAST., Comment: 10 pages, 8 figures, 1 table.Accepted by MNRAS. The raw data used in the article will be published on the FAST website: https://fast.bao.ac.cn. The PID is N2021_4. Please contact the author (liuyao@nao.cas.cn, mz@nao.cas.cn) for processed data