mailitics

不知道 不明了 不想要

  • A Kernelised Stein Discrepancy for Assessing the Fit of Inhomogeneous Random Graph Models

    A Kernelised Stein Discrepancy for Assessing the Fit of Inhomogeneous Random Graph Models arXiv:2505.21580v1 Announce Type: new Abstract: Complex data are often represented as a graph, which in turn can often be viewed as a realisation of a random graph, such as of an inhomogeneous random graph model (IRG). For general fast goodness-of-fit tests in…

  • STACI: Spatio-Temporal Aleatoric Conformal Inference

    STACI: Spatio-Temporal Aleatoric Conformal Inference arXiv:2505.21658v1 Announce Type: new Abstract: Fitting Gaussian Processes (GPs) provides interpretable aleatoric uncertainty quantification for estimation of spatio-temporal fields. Spatio-temporal deep learning models, while scalable, typically assume a simplistic independent covariance matrix for the response, failing to capture the underlying correlation structure. However, spatio-temporal GPs suffer from issues of scalability…

  • Nearly Dimension-Independent Convergence of Mean-Field Black-Box Variational Inference

    Nearly Dimension-Independent Convergence of Mean-Field Black-Box Variational Inference arXiv:2505.21721v1 Announce Type: new Abstract: We prove that, given a mean-field location-scale variational family, black-box variational inference (BBVI) with the reparametrization gradient converges at an almost dimension-independent rate. Specifically, for strongly log-concave and log-smooth targets, the number of iterations for BBVI with a sub-Gaussian family to achieve…

  • Global Minimizers of $ell^p$-Regularized Objectives Yield the Sparsest ReLU Neural Networks

    Global Minimizers of $ell^p$-Regularized Objectives Yield the Sparsest ReLU Neural Networks arXiv:2505.21791v1 Announce Type: new Abstract: Overparameterized neural networks can interpolate a given dataset in many different ways, prompting the fundamental question: which among these solutions should we prefer, and what explicit regularization strategies will provably yield these solutions? This paper addresses the challenge of…

  • A General-Purpose Theorem for High-Probability Bounds of Stochastic Approximation with Polyak Averaging

    A General-Purpose Theorem for High-Probability Bounds of Stochastic Approximation with Polyak Averaging arXiv:2505.21796v1 Announce Type: new Abstract: Polyak-Ruppert averaging is a widely used technique to achieve the optimal asymptotic variance of stochastic approximation (SA) algorithms, yet its high-probability performance guarantees remain underexplored in general settings. In this paper, we present a general framework for establishing…

  • From Data to Stories: Code Agents for KPI Narratives

    From Data to Stories: Code Agents for KPI Narratives HuggingFace’s smolagents framework in action The post From Data to Stories: Code Agents for KPI Narratives appeared first on Towards Data Science. Mariya Mansurova Go to original source

  • Multi-Agent Communication with the A2A Python SDK

    Multi-Agent Communication with the A2A Python SDK The Agent Card helps discover agents, but how does communication between agents actually work in practice? The post Multi-Agent Communication with the A2A Python SDK appeared first on Towards Data Science. Deborah Mesquita Go to original source

  • JAX: Is This Google’s NumPy killer?

    JAX: Is This Google’s NumPy killer? Auto differentiation and JIT compilation make a compelling case. The post JAX: Is This Google’s NumPy killer? appeared first on Towards Data Science. Thomas Reid Go to original source

  • Detecting Malicious URLs Using LSTM and Google’s BERT Models

    Detecting Malicious URLs Using LSTM and Google’s BERT Models A progressive approach to implementing AI-powered webpage detection applications into production The post Detecting Malicious URLs Using LSTM and Google’s BERT Models appeared first on Towards Data Science. Toluwase Babalola Go to original source

  • Tree of Thought Prompting: Teaching LLMs to Think Slowly

    Tree of Thought Prompting: Teaching LLMs to Think Slowly Playing Minesweeper with Augmented Reasoning The post Tree of Thought Prompting: Teaching LLMs to Think Slowly appeared first on Towards Data Science. Shuyang Go to original source

  • 6 types of AI content moderation and how they work

    6 types of AI content moderation and how they work AI will change how organizations moderate content, especially on social media and with the increase in AI-generated content. Here’s what you need to know. Go to techtarget

  • Sisense unveils new suite of AI-powered capabilities

    Sisense unveils new suite of AI-powered capabilities Featuring a natural language interface and autonomous capabilities to augment human analysis, the vendor’s toolkit simplifies developing and embedding advanced applications. Go to techtarget

  • Salesforce to Acquire Informatica for $8 Billion in Equity

    Salesforce to Acquire Informatica for $8 Billion in Equity Salesforce, the world’s leading AI CRM, is acquiring?Informatica, a leader in enterprise AI-powered cloud data management, for approximately $8 billion in equity value, net of Salesforce’s current investment in Informatica. According to the companies, the planned acquisition will enhance Salesforce’s trusted data foundation critical for deploying…

  • Ataccama Accelerates Insights Delivery with Automated Lineage and Cloud-Native Processing

    Ataccama Accelerates Insights Delivery with Automated Lineage and Cloud-Native Processing Ataccama, the data trust company, is releasing Ataccama ONE data trust platform v16.1, introducing powerful data lineage and connectivity capabilities, including enhanced diagram export for audit and compliance use cases and improved lineage visualization tools. Additionally, the platform update also expands pushdown processing for cloud…

  • IBM Officially Closes Acquisition of DataStax

    IBM Officially Closes Acquisition of DataStax DataStax announced its acquisition by IBM is officially closed, allowing the companies to “scale to new heights” and accelerate production AI and NoSQL data at scale. With?Astra DB,?Hyper-Converged Database, and now?watsonx.data, DataStax will provide seamless access to both unstructured and structured data for production AI, according to the company…

  • Precisely Builds on SAP Partnership, Achieves SAP PartnerEdge Build Partner Status

    Precisely Builds on SAP Partnership, Achieves SAP PartnerEdge Build Partner Status Precisely, a global leader in data integrity, is joining the SAP PartnerEdge program as a Build partner, solidifying Precisely’s position as a trusted partner for process automation for SAP solutions and empowering enterprises worldwide to accelerate modernization, streamline complex processes, drive business agility, and…

  • SecurityBridge and Microsoft?Enhance SAP Security With Microsoft Sentinel

    SecurityBridge and Microsoft?Enhance SAP Security With Microsoft Sentinel SecurityBridge, the Cybersecurity Command Center for SAP, is collaborating with Microsoft to integrate SAP data into Microsoft Sentinel?enabling SecurityBridge to seamlessly share SAP security events with Microsoft Sentinel’s cloud-native security information and event management (SIEM). Go to dbta

  • AI-augmented models improve chemical grouting predictions in complex soils

    AI-augmented models improve chemical grouting predictions in complex soils Soil liquefaction—the process where saturated soil loses its structure and transforms to a fluid-like state—can have devastating outcomes, as evidenced by the Great East Japan Earthquake in 2011. Large-scale liquefaction during this disaster damaged thousands of houses in the Tokyo Bay area, posing a formidable challenge…

  • Longer flight delays without compensation? EU plan divides

    Longer flight delays without compensation? EU plan divides The EU is considering allowing airlines to incur longer flight delays without having to compensate passengers in a move that has consumer groups up in arms and is dividing member states. Go to techxplore

  • Telegram to get $300 mn in partnership with Musk’s xAI

    Telegram to get $300 mn in partnership with Musk’s xAI Telegram established a partnership with Elon Musk’s xAI to provide the Grok generative artificial intelligence program on the messaging service for one year, Telegram’s CEO announced Wednesday. Go to techxplore

  • Ultra-thin protective coating boosts cadmium telluride solar cell performance by 13%

    Ultra-thin protective coating boosts cadmium telluride solar cell performance by 13% An NYU Tandon-led research team has developed a novel technique to significantly enhance the performance of cadmium telluride (CdTe) solar cells. Unlike conventional silicon panels that use thick layers of silicon, these solar cells use a simpler, less expensive approach—depositing an ultra-thin layer of…

  • Robot morphs midair to switch from flying to rolling on terrain

    Robot morphs midair to switch from flying to rolling on terrain Specialized robots that can both fly and drive typically touch down on land before attempting to transform and drive away. But when the landing terrain is rough, these robots sometimes get stuck and are unable to continue operating. Go to techxplore

  • Threat actors using aggressive new extortion tactics: report

    Threat actors using aggressive new extortion tactics: report The latest extortion and ransomware report from Palo Alto Networks reveals aggressive new tactics and the escalation of threat actor collaboration. The recently released ‘Unit 42 Extortion and Ransomware Trends January-March 2025’ revealed that threat actors are evolving their tactics, collaborating with state-backed groups and using extortion scams…

  • CommScope launches new fibre termination platform

    CommScope launches new fibre termination platform CommScope has launched a new fibre termination panel platform aimed at enabling simpler upgrades to fibre to the home (FTTH) networks. The CommScope XPND platform comprises a full suite of solutions including panels with interchangeable splice cassettes, adapter modules, optical splitters and cables that can be combined to support…

  • STAT+: HHS cancels nearly $600 million Moderna contract on vaccines for flu pandemics

    STAT+: HHS cancels nearly $600 million Moderna contract on vaccines for flu pandemics This story will be updated The Department of Health and Human Services has notified Moderna that it is canceling a nearly $600 million contract with the company to develop, test, and license vaccines for flu strains that could trigger future pandemics, including…

  • STAT+: PillPack founders’ new health care marketplace has deep roots with Amazon

    STAT+: PillPack founders’ new health care marketplace has deep roots with Amazon A new digital health care marketplace, launched last week, has a good amount of Amazon in its DNA. General Medicine, with $32 million in funding, came out of stealth with three former Amazon employees as co-founders and investors, a business model that could…

  • Opinion: Former FDA commissioner: ‘Cost-cutting’ may undo one of Trump’s best drug pricing achievements

    Opinion: Former FDA commissioner: ‘Cost-cutting’ may undo one of Trump’s best drug pricing achievements President Trump often touted during his first term that his administration had “approved more affordable generic drugs than any administration in history.” He had good reason to highlight these accomplishments. Over the first two years of his presidency, the Food and…

  • Amid measles outbreak, Texas is poised to make vaccine exemptions for kids easier

    Amid measles outbreak, Texas is poised to make vaccine exemptions for kids easier AUSTIN, Texas — Texas this year has been the center of the nation’s largest measles outbreak in more than two decades, as a mostly eradicated disease has sickened more than 700 in the state, sent dozens to hospitals and led to the death of two children…

  • Lilly to acquire biotech developing pain drugs

    Lilly to acquire biotech developing pain drugs Want to stay on top of the science and politics driving biotech today? Sign up to get our biotech newsletter in your inbox. Good morning, we’re seeking nominations for our annual Wunderkinds list, which aims to honor some of the most promising early-career scientists out there. If you have someone in mind, submit…

  • ASIC Sues Former Blockchain Global Exec Over $20M in Unpaid Customer Claims

    ASIC Sues Former Blockchain Global Exec Over $20M in Unpaid Customer Claims ASIC secured interim court orders in February preventing Guo from leaving Australia, but he exited the country days after they expired. Vismaya V Go to decrypt.co

  • Cetus Reveals Recovery Plan, Taps SUI for Bridge Loan

    Cetus Reveals Recovery Plan, Taps SUI for Bridge Loan A Sui community vote requires over 50% participation and majority approval to reclaim $162 million in frozen funds from the Cetus exploit. Vismaya V Go to decrypt.co

  • El Salvador Defies IMF Again With Fresh Bitcoin Purchase Following Loan Review

    El Salvador Defies IMF Again With Fresh Bitcoin Purchase Following Loan Review The IMF urged a halt to crypto accumulation as part of $1.4 billion program. El Salvador responded by buying more. Callan Quinn Go to decrypt.co

  • Ethereum Options Market Signals Cautious Optimism as Open Interest Climbs

    Ethereum Options Market Signals Cautious Optimism as Open Interest Climbs Traders are returning to Ethereum options with split expectations on price targets, even as volatility stays low. Vince Dioquino Go to decrypt.co

  • AI and Crypto Czar David Sacks Says the US Could Buy More Bitcoin

    AI and Crypto Czar David Sacks Says the US Could Buy More Bitcoin AI and crypto czar David Sacks touted the Trump administration’s early wins during Bitcoin 2025, including the pardon of Ross Ulbricht. Jason Nelson Go to decrypt.co

  • Differentially private ratio statistics

    Differentially private ratio statistics arXiv:2505.20351v1 Announce Type: new Abstract: Ratio statistics–such as relative risk and odds ratios–play a central role in hypothesis testing, model evaluation, and decision-making across many areas of machine learning, including causal inference and fairness analysis. However, despite privacy concerns surrounding many datasets and despite increasing adoption of differential privacy, differentially private…

  • Kernel Quantile Embeddings and Associated Probability Metrics

    Kernel Quantile Embeddings and Associated Probability Metrics arXiv:2505.20433v1 Announce Type: new Abstract: Embedding probability distributions into reproducing kernel Hilbert spaces (RKHS) has enabled powerful nonparametric methods such as the maximum mean discrepancy (MMD), a statistical distance with strong theoretical and computational properties. At its core, the MMD relies on kernel mean embeddings to represent distributions…

  • Learning with Expected Signatures: Theory and Applications

    Learning with Expected Signatures: Theory and Applications arXiv:2505.20465v1 Announce Type: new Abstract: The expected signature maps a collection of data streams to a lower dimensional representation, with a remarkable property: the resulting feature tensor can fully characterize the data generating distribution. This “model-free” embedding has been successfully leveraged to build multiple domain-agnostic machine learning (ML)…

  • Covariate-Adjusted Deep Causal Learning for Heterogeneous Panel Data Models

    Covariate-Adjusted Deep Causal Learning for Heterogeneous Panel Data Models arXiv:2505.20536v1 Announce Type: new Abstract: This paper studies the task of estimating heterogeneous treatment effects in causal panel data models, in the presence of covariate effects. We propose a novel Covariate-Adjusted Deep Causal Learning (CoDEAL) for panel data models, that employs flexible model structures and powerful…

  • Balancing Performance and Costs in Best Arm Identification

    Balancing Performance and Costs in Best Arm Identification arXiv:2505.20583v1 Announce Type: new Abstract: We consider the problem of identifying the best arm in a multi-armed bandit model. Despite a wealth of literature in the traditional fixed budget and fixed confidence regimes of the best arm identification problem, it still remains a mystery to most practitioners…

  • Bayesian Optimization for Hyperparameter Tuning of Deep Learning Models

    Bayesian Optimization for Hyperparameter Tuning of Deep Learning Models Explore how Bayesian Optimization outperforms Grid Search in efficiency and performance over binary classification tasks. The post Bayesian Optimization for Hyperparameter Tuning of Deep Learning Models appeared first on Towards Data Science. Kuriko Iwai Go to original source

  • How Microsoft Power BI Elevated My Data Analysis and Visualization Workflow

    How Microsoft Power BI Elevated My Data Analysis and Visualization Workflow Explaining useful features every data analyst needs The post How Microsoft Power BI Elevated My Data Analysis and Visualization Workflow appeared first on Towards Data Science. Benjamin Nweke Go to original source

  • Reinforcement Learning Made Simple: Build a Q-Learning Agent in Python

    Reinforcement Learning Made Simple: Build a Q-Learning Agent in Python Inspired by AlphaGo’s Move 37 — learn how agents explore, exploit, and win The post Reinforcement Learning Made Simple: Build a Q-Learning Agent in Python appeared first on Towards Data Science. Sarah Schürch Go to original source

  • Why Regularization Isn’t Enough: A Better Way to Train Neural Networks with Two Objectives

    Why Regularization Isn’t Enough: A Better Way to Train Neural Networks with Two Objectives Why splitting your objectives and your model might be the key to better performance and clearer trade-offs in deep learning. The post Why Regularization Isn’t Enough: A Better Way to Train Neural Networks with Two Objectives appeared first on Towards Data…

  • How SAP sustainability software helps manage ESG programs

    How SAP sustainability software helps manage ESG programs Gunther Rothermel, who leads engineering for the SAP sustainability line, explains the challenges of regulatory reporting and what AI-infused tools can do to automate the process. Go to techtarget

  • Headless CMS vs. decoupled CMS: What’s the difference?

    Headless CMS vs. decoupled CMS: What’s the difference? Both headless and decoupled CMSes support omnichannel publishing. Yet, headless systems have no native front end, whereas decoupled CMSes have an optional one. Go to techtarget

  • DataOps.live Rolls Out the Dynamic Suite, a Set of Free Snowflake Native Apps

    DataOps.live Rolls Out the Dynamic Suite, a Set of Free Snowflake Native Apps DataOps.live, The Data Products Company, is launching the Dynamic Suite, which includes two new Snowflake Native Apps designed to solve critical data engineering challenges faced by many Snowflake customers: continuous integration and deployment (CI/CD) of Snowflake Objects, and the operationalization of dbt…

  • Operant Woodpecker Offers Open-Source Automated Red Teaming Engine for Kubernetes, APIs, and AI

    Operant Woodpecker Offers Open-Source Automated Red Teaming Engine for Kubernetes, APIs, and AI Operant AI, provider of the Runtime AI Defense Platform, is introducing?Woodpecker, an open-source, automated red teaming engine, that will make advanced security testing accessible to organizations. According to the company, Woodpecker is?designed to help organizations proactively detect and address security vulnerabilities across…

  • Continuent Tungsten v8 Operator for Kubernetes?is Now Available

    Continuent Tungsten v8 Operator for Kubernetes?is Now Available Continuent, a leading provider of solutions for business-critical applications using MySQL and MariaDB databases, is releasing Tungsten version 8 (Tungsten v8) Operator for Kubernetes to deliver a robust, Kubernetes-native solution to simplify and automate the deployment and management of high-availability MySQL clusters.? Go to dbta

  • Exploring the Value of a Time Series Database with InfluxData

    Exploring the Value of a Time Series Database with InfluxData Anais Dotis-Georgiou, product manager, InfluxData, joined DBTA’s webinar, Why a Time Series Database is the Best Solution for Building Real-Time, Intelligent Systems and Applications, to discuss the scenarios in which time series databases rival traditional relational databases, offering a variety of benefits relating to performance,…

  • UK loot box self-regulation fails: New study finds rampant non-compliance and no enforcement

    UK loot box self-regulation fails: New study finds rampant non-compliance and no enforcement Loot boxes and gacha are gambling-like products inside video games that players buy to obtain random rewards. Concerns have been raised about consumers, particularly children, experiencing financial harm and developing gambling problems. The previous Conservative UK government asked the industry, represented by…

  • Ban fossil fuel heating systems? A way out of the war of beliefs

    Ban fossil fuel heating systems? A way out of the war of beliefs In several industrialized countries, governments are backing away from controversial building energy legislation that sought to ban oil and gas heating and replace them with fossil-free systems. Go to techxplore

  • Smart measures to reduce your electricity bill

    Smart measures to reduce your electricity bill Would you adjust your electricity consumption if you received a notification on your mobile phone telling you when electricity was going to be most expensive the following day? Research shows that good information can influence our energy consumption. Go to techxplore

  • Q&A: Multimodality as the next big leap for AI

    Q&A: Multimodality as the next big leap for AI As the head of the Natural Language Processing Laboratory at EPFL, Antoine Bosselut keeps a close eye on the development of generative artificial intelligence tools such as ChatGPT. He looks back at their evolution over the past two years and suggests some avenues for the future.…

  • Dehydration warning at your fingertips: Touchscreen tech tracks body water levels

    Dehydration warning at your fingertips: Touchscreen tech tracks body water levels The holy month of Ramadan is a sacred time when millions of Muslims around the world embark on a profound spiritual journey of fasting, prayer, and reflection. But it is also a time when many face serious health risks, as going without food or…

  • Kyndryl expands Skytap platform to Australia

    Kyndryl expands Skytap platform to Australia Mission-critical technology services provider Kyndryl has launched its cloud modernisation solution Skytap into the Australian market. The solution, which aims to help businesses migrate business-critical applications to the cloud, is now available to customers in the Microsoft Azure Australia East data centre region in Sydney. The deployment marks the…

  • How agentic AI will revolutionise customer experience in Australia

    How agentic AI will revolutionise customer experience in Australia The COVID-19 pandemic changed the way Australian businesses interact with their customers. Faced with lockdowns, staff shortages and a massive shift to digital channels, many turned to automation as a lifeline. Chatbots — quick to deploy and relatively cost-effective — were the obvious choice. But what…

  • AEMC drawing up new electricity connection rules to manage AI boom

    AEMC drawing up new electricity connection rules to manage AI boom The Australian Energy Market Commission (AEMC) has announced it has completed a comprehensive overhaul of the technical requirements for connecting to the national electricity grid, which includes addressing emerging challenges to the grid from new large energy users such as data centres and hydrogen…

  • How to prepare your data for AI success

    How to prepare your data for AI success AI promises business transformation, but those efforts are likely to fail without the right data foundations. According to Gartner, 63% of organisations either lack or are unsure if they have the right data management practices needed to support AI. As a result, Gartner predicts 60% of AI…

  • STAT+: Lawmaker wants to know how FDA can police drug ads after cutting its oversight workforce

    STAT+: Lawmaker wants to know how FDA can police drug ads after cutting its oversight workforce Amid ongoing controversy over pharmaceutical advertising, one lawmaker wants to know how the U.S. government will enforce regulations after the Food and Drug Administration let go of numerous employees from the office that oversees prescription drug promotions. In a…

  • STAT+: Young scientists say they may abandon research as their career options shrink amid Trump cuts

    STAT+: Young scientists say they may abandon research as their career options shrink amid Trump cuts Becks Padrusch‘s fondest memories growing up were of trips to Boston’s Museum of Science, where the Arlington native got to touch animal organs and watch with fascination as chickens hatched in incubators. As a toddler, Padrusch, who uses they/them pronouns,…

  • STAT+: What’s already been lost from CDC layoffs

    STAT+: What’s already been lost from CDC layoffs You’re reading the web edition of D.C. Diagnosis, STAT’s twice-weekly newsletter about the politics and policy of health and medicine. Sign up here to receive it in your inbox on Tuesdays and Thursdays. Summer holiday weekends are a challenge. Do I seek a “relaxing” long weekend sitting in traffic…

  • RFK Jr. rolls back Covid vaccine recommendations for healthy children, pregnant people

    RFK Jr. rolls back Covid vaccine recommendations for healthy children, pregnant people Health secretary Robert F. Kennedy Jr. announced Tuesday that he has unilaterally struck the recommendation that healthy children and healthy pregnant people get Covid-19 booster shots — a move that experts say is unprecedented. Kennedy made the announcement on the social media site X,…

  • Jupiter Price Surges Amid Expansion and Bitcoin Tailwinds

    Jupiter Price Surges Amid Expansion and Bitcoin Tailwinds Jupiter’s native token JUP’s rally points to deeper momentum across Solana and macro markets, analysts told Decrypt. Vismaya V Go to decrypt.co

  • Pakistan Appoints World Liberty Financial Advisor to Key Government Role on Crypto

    Pakistan Appoints World Liberty Financial Advisor to Key Government Role on Crypto Bilal Bin Saqib has been tapped to lead strategy as Islamabad embraces crypto mining and ties with a controversial U.S. crypto project. Callan Quinn Go to decrypt.co

  • Civitai Turns to Crypto After Credit Card Processor Ban Over AI Explicit Content

    Civitai Turns to Crypto After Credit Card Processor Ban Over AI Explicit Content AI art platform Civitai now accepts eight cryptos to purchase its own virtual currency, joining other NSFW businesses embracing the tech. Jose Antonio Lanz Go to decrypt.co

  • First Humanoid Boxing Match Takes Place in China—And It’s Pretty Cool

    First Humanoid Boxing Match Takes Place in China—And It’s Pretty Cool Four Unitree robots battled it out in a fighting competition, controlled by human operators in an event reminiscent of 2011’s Real Steel. Ryan Gladwin Go to decrypt.co

  • This Project Lets You Send Bitcoin Without Internet Access

    This Project Lets You Send Bitcoin Without Internet Access A pseudonymous developer has unveiled a hackathon project enabling Bitcoin transactions without internet access via long-range radio. Simon Chandler Go to decrypt.co

  • Preconditioned Langevin Dynamics with Score-Based Generative Models for Infinite-Dimensional Linear Bayesian Inverse Problems

    Preconditioned Langevin Dynamics with Score-Based Generative Models for Infinite-Dimensional Linear Bayesian Inverse Problems arXiv:2505.18276v1 Announce Type: new Abstract: Designing algorithms for solving high-dimensional Bayesian inverse problems directly in infinite-dimensional function spaces – where such problems are naturally formulated – is crucial to ensure stability and convergence as the discretization of the underlying problem is refined.…

  • Operator Learning for Schr”{o}dinger Equation: Unitarity, Error Bounds, and Time Generalization

    Operator Learning for Schr”{o}dinger Equation: Unitarity, Error Bounds, and Time Generalization arXiv:2505.18288v1 Announce Type: new Abstract: We consider the problem of learning the evolution operator for the time-dependent Schr”{o}dinger equation, where the Hamiltonian may vary with time. Existing neural network-based surrogates often ignore fundamental properties of the Schr”{o}dinger equation, such as linearity and unitarity, and…

  • Online Statistical Inference of Constrained Stochastic Optimization via Random Scaling

    Online Statistical Inference of Constrained Stochastic Optimization via Random Scaling arXiv:2505.18327v1 Announce Type: new Abstract: Constrained stochastic nonlinear optimization problems have attracted significant attention for their ability to model complex real-world scenarios in physics, economics, and biology. As datasets continue to grow, online inference methods have become crucial for enabling real-time decision-making without the need…

  • On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspective

    On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspective arXiv:2505.18346v1 Announce Type: new Abstract: Weak-to-strong generalization, where a student model trained on imperfect labels generated by a weaker teacher nonetheless surpasses that teacher, has been widely observed but the mechanisms that enable it have remained poorly understood. In this paper, through a theoretical analysis of…

  • Identifiability of latent causal graphical models without pure children

    Identifiability of latent causal graphical models without pure children arXiv:2505.18410v1 Announce Type: new Abstract: This paper considers a challenging problem of identifying a causal graphical model under the presence of latent variables. While various identifiability conditions have been proposed in the literature, they often require multiple pure children per latent variable or restrictions on the…

  • Code Agents: The Future of Agentic AI

    Code Agents: The Future of Agentic AI HuggingFace smolagents framework in action The post Code Agents: The Future of Agentic AI appeared first on Towards Data Science. Mariya Mansurova Go to original source

  • How to Reduce Your Power BI Model Size by 90%

    How to Reduce Your Power BI Model Size by 90% Have you ever wondered what makes Power BI so fast and powerful when it comes to performance? Learn on a real-life example about data model optimization and general rules for reducing data model The post How to Reduce Your Power BI Model Size by 90%…

  • The Best AI Books & Courses for Getting a Job

    The Best AI Books & Courses for Getting a Job A comprehensive guide to the books and courses that helped me learn AI The post The Best AI Books & Courses for Getting a Job appeared first on Towards Data Science. Egor Howell Go to original source

  • How to Generate Synthetic Data: A Comprehensive Guide Using Bayesian Sampling and Univariate Distributions

    How to Generate Synthetic Data: A Comprehensive Guide Using Bayesian Sampling and Univariate Distributions Data makes the engine run in many organisations. But what if the number of observations is too low or there is only expert knowledge? I will demonstrate how to generate synthetic data with applications in predictive maintenance. The post How to…

  • Understanding Matrices | Part 1: Matrix-Vector Multiplication

    Understanding Matrices | Part 1: Matrix-Vector Multiplication The physical meaning of multiplying a matrix by a vector, and how it works on several special matrices. The post Understanding Matrices | Part 1: Matrix-Vector Multiplication appeared first on Towards Data Science. Tigran Hayrapetyan Go to original source

  • Novel elastic alloy achieves 20x temperature change and 90% Carnot efficiency in solid-state heat pumping

    Novel elastic alloy achieves 20x temperature change and 90% Carnot efficiency in solid-state heat pumping Researchers at the School of Engineering of the Hong Kong University of Science and Technology (HKUST) have developed a novel elastic alloy called Ti78Nb22, which achieves remarkable efficiency for solid-state heat pumping and exhibits a reversible temperature change (ΔT) ability…

  • Proposed wave energy park could generate power while shielding Portuguese coastline

    Proposed wave energy park could generate power while shielding Portuguese coastline A study by researchers from the Interdisciplinary Center of Marine and Environmental Research (CIIMAR) and the Faculty of Engineering of the University of Porto (FEUP) analyzes the potential and feasibility of a wave energy converter park off the coast of Esposende, Portugal, with the…

  • Self-trained vision transformers mimic human gaze with surprising precision

    Self-trained vision transformers mimic human gaze with surprising precision Can machines ever see the world as we see it? Researchers have uncovered compelling evidence that vision transformers (ViTs), a type of deep-learning model that specializes in image analysis, can spontaneously develop human-like visual attention patterns when trained without labeled instructions. Go to techxplore

  • AI model pinpoints sources of driver stress, paving the way for smart driving assistants

    AI model pinpoints sources of driver stress, paving the way for smart driving assistants In 2024, 1,040 accidents were recorded on Spanish roads, in addition to minor collisions and other driving problems. The causes of these accidents include speeding, adverse weather conditions and substance abuse, but also distraction and stressful situations that can be mitigated…

  • Tool automatically separates training and test data to improve AI evaluation

    Tool automatically separates training and test data to improve AI evaluation A new tool has been developed to better assess the performance of AI models. It was developed by bioinformaticians at Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU) and the Helmholtz Institute for Pharmaceutical Research Saarland (HIPS). Go to techxplore

  • STAT+: Harvard is the most celebrated university in the world. Will Trump’s international student ban derail that standing?

    STAT+: Harvard is the most celebrated university in the world. Will Trump’s international student ban derail that standing? Harvard University has long stood on top of the academic world. Thousands from all over seek a coveted spot every year at the Ivy League school with a reputation for cultivating future leaders, from Nobel Prize recipients to…

  • STAT+: What Baby KJ means for the CRISPR gene editing industry

    STAT+: What Baby KJ means for the CRISPR gene editing industry For the ailing gene editing industry, hope came earlier this month in the tiny, smiling, fuzzy-headed form of KJ Muldoon. At just 6 months old, KJ received a gene editing treatment custom-built to correct his unique mutation. He’s not cured, researchers explained at the…

  • Rich Dad Poor Dad Author Can’t Believe People Aren’t Buying Bitcoin

    Rich Dad Poor Dad Author Can’t Believe People Aren’t Buying Bitcoin Kiyosaki says even 0.01 BTC could make you rich in two years, calling Bitcoin the “easiest” path to wealth. Vismaya V Go to decrypt.co

  • How the Crypto Industry Is Responding to the CFTC’s Call on Perpetuals

    How the Crypto Industry Is Responding to the CFTC’s Call on Perpetuals Some of crypto’s biggest players have weighed in on the CFTC’s April request. Here’s how they think the U.S. can bring perpetuals home. Vince Dioquino Go to decrypt.co

  • Ex-NFL Star Tom Brady Returns to Crypto With Investment in AI Startup

    Ex-NFL Star Tom Brady Returns to Crypto With Investment in AI Startup Despite losses stemming from FTX’s collapse, Brady is getting back into crypto. This time, he’s investing in an AI-native fintech startup. Vince Dioquino Go to decrypt.co

  • Crypto Investor Charged With Kidnapping, Torturing Man in NYC Over Bitcoin Password

    Crypto Investor Charged With Kidnapping, Torturing Man in NYC Over Bitcoin Password The New York assault is the latest addition in a growing list of violent attempts to steal crypto through physical coercion. Callan Quinn Go to decrypt.co

  • Bitcoin Rebounds as Trump Extends EU Tariff Deadline, US Futures Tick Higher

    Bitcoin Rebounds as Trump Extends EU Tariff Deadline, US Futures Tick Higher Bitcoin has climbed back above $109,600 as Trump delays EU tariffs, easing trade tensions and fueling renewed optimism across risk assets. Sebastian Sinclair Go to decrypt.co

  • Learning Probabilities of Causation from Finite Population Data

    Learning Probabilities of Causation from Finite Population Data arXiv:2505.17133v1 Announce Type: new Abstract: Probabilities of causation play a crucial role in modern decision-making. This paper addresses the challenge of predicting probabilities of causation for subpopulations with textbf{insufficient} data using machine learning models. Tian and Pearl first defined and derived tight bounds for three fundamental probabilities…

  • Liouville PDE-based sliced-Wasserstein flow for fair regression

    Liouville PDE-based sliced-Wasserstein flow for fair regression arXiv:2505.17204v1 Announce Type: new Abstract: The sliced Wasserstein flow (SWF), a nonparametric and implicit generative gradient flow, is applied to fair regression. We have improved the SWF in a few aspects. First, the stochastic diffusive term from the Fokker-Planck equation-based Monte Carlo is transformed to Liouville partial differential…

  • Deconfounded Warm-Start Thompson Sampling with Applications to Precision Medicine

    Deconfounded Warm-Start Thompson Sampling with Applications to Precision Medicine arXiv:2505.17283v1 Announce Type: new Abstract: Randomized clinical trials often require large patient cohorts before drawing definitive conclusions, yet abundant observational data from parallel studies remains underutilized due to confounding and hidden biases. To bridge this gap, we propose Deconfounded Warm-Start Thompson Sampling (DWTS), a practical approach…

  • Learning to Choose or Choosing to Learn: Best-of-N vs. Supervised Fine-Tuning for Bit String Generation

    Learning to Choose or Choosing to Learn: Best-of-N vs. Supervised Fine-Tuning for Bit String Generation arXiv:2505.17288v1 Announce Type: new Abstract: Using the bit string generation problem as a case study, we theoretically compare two standard methods for adapting large language models to new tasks. The first, referred to as supervised fine-tuning, involves training a new…

  • Optimal Transport with Heterogeneously Missing Data

    Optimal Transport with Heterogeneously Missing Data arXiv:2505.17291v1 Announce Type: new Abstract: We consider the problem of solving the optimal transport problem between two empirical distributions with missing values. Our main assumption is that the data is missing completely at random (MCAR), but we allow for heterogeneous missingness probabilities across features and across the two distributions.…

  • Weekly Entering & Transitioning – Thread 26 May, 2025 – 02 Jun, 2025

    Weekly Entering & Transitioning – Thread 26 May, 2025 – 02 Jun, 2025 Welcome to this week’s entering & transitioning thread! This thread is for any questions about getting started, studying, or transitioning into the data science field. Topics include: Learning resources (e.g. books, tutorials, videos) Traditional education (e.g. schools, degrees, electives) Alternative education (e.g.…

  • 2025 stack check: which DS/ML tools am I missing?

    2025 stack check: which DS/ML tools am I missing? Hi all, I work in ad-tech, where my job is to improve the product with data-driven algorithms, mostly on tabular datasets (CTR models, bidding, attribution, the usual). Current work stack (quite classic I guess) pandas, numpy, scikit-learn, xgboost, statsmodels PyTorch (light use) JupyterLab & notebooks matplotlib,…

  • Can you explain to me the product analytics job?

    Can you explain to me the product analytics job? I ve watched videos about Data Scientist Product Analytics but i still dont understand if the job would excite me. Can someone explain it more in depth so that i can understand if i like it? I like the data science job (i am pursuing a…

  • Found a really amazing video , providing context to the breakthrough as well as the misconceived hype around Alphaevolve

    Found a really amazing video , providing context to the breakthrough as well as the misconceived hype around Alphaevolve I am sure by now most of us would have seen or atleast heard about AlphaEvolve and it’s many breakthroughs including the 4*4 MM improvement. While this was a fantastic step forward in constrained optimisation problems…

  • Is studying Data Science still worth it?

    Is studying Data Science still worth it? Hi everyone, I’m currently studying data science, but I’ve been hearing that the demand for data scientists is decreasing significantly. I’ve also been told that many data scientists are essentially becoming analysts, while the machine learning side of things is increasingly being handled by engineers. Does it still…