AI Research
[2506.23629] A Nonlinear Low-rank Representation Model with Convolutional Neural Network for Imputing Water Quality Data

View a PDF of the paper titled A Nonlinear Low-rank Representation Model with Convolutional Neural Network for Imputing Water Quality Data, by Xin Liao and Bing Yang and Cai Yu
Abstract:The integrity of Water Quality Data (WQD) is critical in environmental monitoring for scientific decision-making and ecological protection. However, water quality monitoring systems are often challenged by large amounts of missing data due to unavoidable problems such as sensor failures and communication delays, which further lead to water quality data becoming High-Dimensional and Sparse (HDS). Traditional data imputation methods are difficult to depict the potential dynamics and fail to capture the deep data features, resulting in unsatisfactory imputation performance. To effectively address the above issues, this paper proposes a Nonlinear Low-rank Representation model (NLR) with Convolutional Neural Networks (CNN) for imputing missing WQD, which utilizes CNNs to implement two ideas: a) fusing temporal features to model the temporal dependence of data between time slots, and b) Extracting nonlinear interactions and local patterns to mine higher-order relationships features and achieve deep fusion of multidimensional information. Experimental studies on three real water quality datasets demonstrate that the proposed model significantly outperforms existing state-of-the-art data imputation models in terms of estimation accuracy. It provides an effective approach for handling water quality monitoring data in complex dynamic environments.
Submission history
From: Xin Liao [view email]
[v1]
Mon, 30 Jun 2025 08:48:19 UTC (411 KB)
[v2]
Wed, 10 Sep 2025 12:50:14 UTC (411 KB)
AI Research
YSU: Grant puts YSU at forefront of AI research – WFMJ.com
AI Research
Advarra launches AI- and data-backed study design solution to improve operational efficiency in clinical trials

Advarra, the market leader in regulatory reviews and a leading provider of clinical research technology, today announced the launch of its Study Design solution, which uses AI- and data-driven insights to help life sciences companies design protocols for greater operational efficiency in the real world.
Study Design solution evaluates a protocol’s feasibility by comparing it to similar trials using Braid™, Advarra’s newly launched data and AI engine. Braid is powered by a uniquely rich set of digitized protocol-related documents and operational data from over 30,000 historical studies conducted by 3,500 sponsors. Drawing on Advarra’s institutional review board (IRB) and clinical trial systems, this dataset spans diverse trial types and therapeutic areas, provides granular detail on schedules of assessment, and tracks longitudinal study modifications, giving sponsors deeper insights than solutions based only on in-house or public datasets.
“Too often, clinical trial protocols are developed without the benefit of robust comparative intelligence, leading to inefficient designs and operations,” said Laura Russell, senior vice president, head of data and AI product development at Advarra. “By drawing on the industry’s largest and richest operational dataset, Advarra’s Study Design solution delivers deeper insights into the feasibility of a protocol’s design. It helps sponsors better anticipate downstream operational challenges, make more informed decisions to simplify trial designs, and accelerate protocol development timelines.”
Advarra’s Study Design solution can be used to optimize a protocol prior to final submission or for retrospective analyses. The solution provides insights on design factors that drive operational feasibility, such as the impact of eligibility criteria, burdensomeness of the schedule of assessment on sites and participants, and reasons for amendments. Study teams receive custom benchmarking that allows for operational risk assessments through tailored data visualizations and consultations with Advarra’s data and study design experts. Technical teams can work directly within Advarra’s secure, self-service insights workspace to explore operational data for the purpose of powering internal analyses, models, and business intelligence tools.
“Early pilots have already demonstrated measurable impact,” added Russell. “In one engagement, benchmarking a sponsor’s protocol against comparable studies revealed twice as many exclusion criteria and 60 percent more site visits than industry benchmarks. With these insights, the sponsor saw a path to streamline future trial designs by removing unnecessary criteria, clustering procedures, and adopting hybrid visit models, ultimately reducing site burden and making participation easier for patients.”
Study Design solution is the first in a series of offerings by Advarra that will be powered by Braid. Future applications will extend insights beyond protocol design to improve study startup, enhance collaboration, and better support sites.
To learn more about Study Design solution or to request a consultation, visit advarra.com/study-design.
About Advarra
Advarra breaks the silos that impede clinical research, aligning patients, sites, sponsors, and CROs in a connected ecosystem to accelerate trials. Advarra is number one in research review services, a leader in site and sponsor technology, and is trusted by the top 50 global biopharma sponsors, top 20 CROs, and 50,000 site investigators worldwide. Advarra solutions enable collaboration, transparency, and speed to optimize trial operations, ensure patient safety and engagement, and reimagine clinical research while improving compliance. For more information, visit advarra.com.
AI Research
Best Artificial Intelligence (AI) Stock to Buy Now: Nvidia or Palantir?

Palantir has outperformed Nvidia so far this year, but investors shouldn’t ignore the chipmaker’s valuation.
Artificial intelligence (AI) investing is a remarkably broad field, as there are numerous ways to profit from this trend. Two of the most popular are Nvidia (NVDA -1.55%) and Palantir (PLTR -0.58%), which represent two different sides of AI investing.
Nvidia is on the hardware side, while Palantir produces AI software. These are two lucrative fields to invest in, but is there a clear-cut winner? Let’s find out.
Image source: Getty Images.
Palantir’s business model is more sustainable
Nvidia manufactures graphics processing units (GPUs), which have become the preferred computing hardware for processing AI workloads. While Nvidia has made a ton of money selling GPUs, it’s not done yet. Nvidia expects the big four AI hyperscalers to spend around $600 billion in data center capital expenditures this year, but projects that global data center capital expenditures will increase to $3 trillion to $4 trillion by 2030. That’s a major spending boom, and Nvidia will reap a substantial amount of money from that rise.
However, Nvidia isn’t completely safe. Its GPUs could fall out of style with AI hyperscalers as they develop in-house AI processing chips that could steal some of Nvidia’s market share. Furthermore, if demand for computing equipment diminishes, Nvidia’s revenue streams could fall. That’s why a subscription model like Palantir is a better business over the long term.
Palantir develops AI software that can be described as “data in, insights out.” By using AI to process a ton of information rapidly, Palantir can provide real-time insights for what those with decision-making authority should do. Furthermore, it also gives developers the power to deploy AI agents, which can act autonomously within a business.
Palantir sells its software to commercial clients and government entities, and has gathered a sizable customer base, although that figure is rapidly expanding. As the AI boom continues, these customers will likely stick with Palantir because it’s incredibly difficult to move away from the software once it has been deployed. This means that after the AI spending boom is complete, Palantir will still be able to generate continuous revenue from its software subscriptions.
This gives Palantir a business advantage.
Nvidia is growing faster
Although Palantir’s revenue growth is accelerating, it’s still slower than Nvidia’s.
NVDA Revenue (Quarterly YoY Growth) data by YCharts
This may invert sometime in the near future, but for now, Nvidia has the growth edge.
One item that could reaccelerate Nvidia’s growth is the return of its business in China. Nvidia is currently working on obtaining its export license for H20 chips. Once that is returned, the company could see a massive demand from another country that requires significant AI computing power. Even without a massive chunk of sales, Nvidia is still growing faster than Palantir, giving it the advantage here.
Nvidia is far cheaper than Palantir
With both companies growing at a similar rate, it would be logical to expect that they should trade within a similar valuation range. However, that’s not the case. Whether you analyze the stocks from a forward price-to-earnings (P/E) or price-to-sales (P/S) basis, Palantir’s stock is unbelievably expensive.
NVDA PE Ratio (Forward) data by YCharts
From a P/S basis, Palantir is about 5 times more expensive than Nvidia. From a forward P/E basis, it’s about 6.5 times more expensive.
With these two growing at the same rate, this massive premium for Palantir’s stock doesn’t make a ton of sense. It will take years, or even a decade, at Palantir’s growth rate to bring its valuation down to a reasonable level; yet, Nvidia is already trading at that price point.
I think this gives Nvidia an unassailable advantage for investors, and I think it’s the far better buy right now, primarily due to valuation, as Palantir’s price has gotten out of control.
Keithen Drury has positions in Nvidia. The Motley Fool has positions in and recommends Nvidia and Palantir Technologies. The Motley Fool has a disclosure policy.
-
Business3 weeks ago
The Guardian view on Trump and the Fed: independence is no substitute for accountability | Editorial
-
Tools & Platforms1 month ago
Building Trust in Military AI Starts with Opening the Black Box – War on the Rocks
-
Ethics & Policy2 months ago
SDAIA Supports Saudi Arabia’s Leadership in Shaping Global AI Ethics, Policy, and Research – وكالة الأنباء السعودية
-
Events & Conferences4 months ago
Journey to 1000 models: Scaling Instagram’s recommendation system
-
Jobs & Careers3 months ago
Mumbai-based Perplexity Alternative Has 60k+ Users Without Funding
-
Podcasts & Talks2 months ago
Happy 4th of July! 🎆 Made with Veo 3 in Gemini
-
Education2 months ago
Macron says UK and France have duty to tackle illegal migration ‘with humanity, solidarity and firmness’ – UK politics live | Politics
-
Education3 months ago
VEX Robotics launches AI-powered classroom robotics system
-
Podcasts & Talks2 months ago
OpenAI 🤝 @teamganassi
-
Funding & Business3 months ago
Kayak and Expedia race to build AI travel agents that turn social posts into itineraries