Category Archives: Data Science

Gilbane Building Company successfully completes construction of the University of Virginia’s cutting-edge School of … – PR Web

"This beautiful and unique new space gives students, faculty, and staff a home base for research and teaching that intersects with schools and disciplines across Grounds and beyond," said University President Jim Ryan.

This four-story, 61,000-square-foot facility for the School of Data Science embodies the core principles of openness and collaboration that drive the school's mission. Designed with open hallways, a spacious atrium, monumental stairs, adaptive classrooms, and research and meeting areas, the building promotes a culture of innovation and teamwork.

"We are delighted to continue our partnership with UVA on this cutting-edge facility that creates opportunity for collaborative, open, and responsible data science research and education as the school's goals state," said Maggie Reed, Gilbane's Richmond, Virginia, Business Leader. "While we helped build a physical building to house the program, we are excited about what this school without walls will create in the future."

An interactive data sculpture, designed by SoSo Limited and built by Hypersonic, will be a unique feature of the building, allowing visitors to engage with data sets and explore various data stories. The sculpture, hanging from a skylight at the building's core, will be visible from the street and every floor.

Inclusivity is a priority, and the facility features a wellness space, a lactation space, and non-gendered restrooms, making it a welcoming environment for all.

Sustainability is a crucial focus, evident in the LEED Gold-certified green building design. Solar panels, daylighting strategies, and shading systems reduce the need for artificial lighting. The School of Data Science will receive 15% of its power from solar energy provided by four arrays of solar photovoltaic panels installed on the roof of its soon-to-open new home. This aligns with UVA's Sustainability Plan, aiming for carbon neutrality by 2030 and fossil fuel-free status by 2050.

About the University of Virginia UVA is an iconic public institution of higher education, boasting nationally ranked schools and programs, diverse and distinguished faculty, a major academic medical center, and a proud history as a renowned research university. The community and culture of the University are enriched by active student self-governance, sustained commitment to the arts, and a robust NCAA Division I Athletics program. In its third century, the University of Virginia offers an affordable, world-class education consistently ranked among the nation's best. As one of the nation's leading public institutions, UVA pushes the boundaries of what's possible always in the name of the greater good. One of the things that makes this possible is an unswerving commitment to initiatives that grow, strengthen, and shape our institution for the future. For more information visit: https://www.virginia.edu.

About Gilbane Building Company Gilbane provides a full slate of construction and facilities-related services from preconstruction services planning and integrated consulting capabilities to comprehensive construction management, general contracting, design-build, and facility management services for clients across various markets. Founded in 1870 and still a privately held, family-owned company, Gilbane has more than 45 office locations worldwide. For more information, visit http://www.gilbaneco.com.

About Gilbane Richmond Serving the Richmond community for over 50 years as a construction industry leader, Gilbane has the breadth of experience and local knowledge to partner with corporations, institutions, schools and universities, government agencies, and attractions in Virginia. Gilbane has constructed many of Richmond's local iconic landmarks and facilities, including the Virginia State Capitol Building Restoration and Extension, the Altria Corporate Headquarters Campus, the Virginia Commonwealth University Institute of Contemporary Art, CenterStage and Landmark Theatre, the Altria Theater Renovations as well as several major renovation projects for Capitol One on their West Creek campus.

Media Contact

Heidi K. Bodine, Gilbane Building Company, 407-204-4023, [emailprotected], http://www.gilbaneco.com

SOURCE Gilbane Building Company

Continued here:

Gilbane Building Company successfully completes construction of the University of Virginia's cutting-edge School of ... - PR Web

Data Scientist Breakdown: Skills, Certifications, and Salary – KDnuggets

A lot of us are worried about the demand for data scientists since the use of platforms such as ChatGPT. Over the past few years, companies have been laying off employees in the tech sector, and the big question everybody is asking is whether AI is the reason behind it.

In today's article, we will be speaking specifically about data science, and although there are challenges, those with data science skills have a more promising career longevity.

A study by 365datascience shows that data scientists made up 3% of those laid off by major tech companies. Other tech professionals such as software engineers, were affected more, at around 22%.

This statistic alone presents to us the crucial role data scientists play in advancing the tech industry.

A data scientist's job role focuses on statistics, machine learning, and artificial intelligence. Their business objective is to be able to use different data strategies and turn raw data into business insights that can be used in the decision-making process.

This can go from simple data analysis to building machine learning models.

A data scientist is skilled in mathematics, statistics, and computer science with expertise in a programming language such as Python or R.

As stated above, to become a data scientist, you will need to have a good foundational understanding of mathematics, and statistics along with a programming language.

What about computer science? Do I not need a degree for this?

In some cases yes, depending on where you are in the world. For example, in the UK a lot of companies desire a university degree. However, as the demand for data scientists continues to grow, organisations understand the low supply and are more than happy to take on people with the correct certifications and skills.

So what kind of certifications are these?

So whats the money like?

According to Glassdoor, updated on the 12th of April 2024 - the average salary for a data scientist in the US is $157,000, ranging from $132,00 to $190,000.

Please note that in this figure, only 37.8% of job postings announced their salary. Working in the tech industry, I have come across US data scientists with a salary between $160,000$200,000 annually.

However, salary is highly dependent on a range of factors:

The demand for data scientists will continue to grow and if you are somebody who is looking to transition into the tech industry with a career that has a higher chance of job security - data science is for you.

Dont worry about not having the right qualifications from University, you can gain the same experience, skills, and land a job with the certifications mentioned above!

Nisha Arya is a data scientist, freelance technical writer, and an editor and community manager for KDnuggets. She is particularly interested in providing data science career advice or tutorials and theory-based knowledge around data science. Nisha covers a wide range of topics and wishes to explore the different ways artificial intelligence can benefit the longevity of human life. A keen learner, Nisha seeks to broaden her tech knowledge and writing skills, while helping guide others.

The rest is here:

Data Scientist Breakdown: Skills, Certifications, and Salary - KDnuggets

The Landscape of Emerging AI Agent Architectures for Reasoning, Planning, and Tool Calling: A – Towards Data Science

Since the launch of ChatGPT, the initial wave of generative AI applications has largely revolved around chatbots that utilize the Retrieval Augmented Generation (RAG) pattern to respond to user prompts. While there is ongoing work to enhance the robustness of these RAG-based systems, the research community is now exploring the next generation of AI applications a common theme being the development of autonomous AI agents.

Agentic systems incorporate advanced capabilities like planning, iteration, and reflection, which leverage the models inherent reasoning abilities to accomplish tasks end-to-end. Paired with the ability to use tools, plugins, and function calls agents are empowered to tackle a wider range of general-purpose work.

Reasoning is a foundational building block of the human mind. Without reasoning one would not be able to make decisions, solve problems, or refine plans when new information is learned essentially misunderstanding the world around us. If agents dont have strong reasoning skills then they might misunderstand their task, generate nonsensical answers, or fail to consider multi-step implications.

We find that most agent implementations contain a planning phase which invokes one of the following techniques to create a plan: task decomposition, multi-plan selection, external module-aided planning, reflection and refinement and memory-augmented planning [1].

Another benefit of utilizing an agent implementation over just a base language model is the agents ability to solve complex problems by calling tools. Tools can enable an agent to execute actions such as interacting with APIs, writing to third party applications, and more. Reasoning and tool calling are closely intertwined and effective tool calling has a dependency on adequate reasoning. Put simply, you cant expect an agent with poor reasoning abilities to understand when is the appropriate time to call its tools.

Our findings emphasize that both single-agent and multi-agent architectures can be used to solve challenging tasks by employing reasoning and tool calling steps.

For single agent implementations, we find that successful goal execution is contingent upon proper planning and self-correction [1, 2, 3, 4]. Without the ability to self-evaluate and create effective plans, single agents may get stuck in an endless execution loop and never accomplish a given task or return a result that does not meet user expectations [2]. We find that single agent architectures are especially useful when the task requires straightforward function calling and does not need feedback from another agent.

However, we note that single agent patterns often struggle to complete a long sequence of sub tasks or tool calls [5, 6]. Multi-agent patterns can address the issues of parallel tasks and robustness since multiple agents within the architecture can work on individual subproblems. Many multi-agent patterns start by taking a complex problem and breaking it down into several smaller tasks. Then, each agent works independently on solving each task using their own independent set of tools.

Architectures involving multiple agents present an opportunity for intelligent labor division based on capabilities as well as valuable feedback from diverse agent personas. Numerous multi-agent architectures operate in stages where teams of agents are dynamically formed and reorganized for each planning, execution, and evaluation phase [7, 8, 9]. This reorganization yields superior outcomes because specialized agents are utilized for specific tasks and removed when no longer required. By matching agent roles and skills to the task at hand, agent teams can achieve greater accuracy and reduce the time needed to accomplish the goal. Crucial features of effective multi-agent architectures include clear leadership within agent teams, dynamic team construction, and efficient information sharing among team members to prevent important information from getting lost amidst superfluous communication.

Our research highlights notable single agent methods such as ReAct, RAISE, Reflexion, AutoGPT + P, LATS, and multi agent implementations such as DyLAN, AgentVerse, and MetaGPT, which are explained more in depth in the full text.

Single Agent Patterns:

Single agent patterns are generally best suited for tasks with a narrowly defined list of tools and where processes are well-defined. They dont face poor feedback from other agents or distracting and unrelated chatter from other team members. However, single agents may get stuck in an execution loop and fail to make progress towards their goal if their reasoning and refinement capabilities arent robust.

Multi Agent Patterns:

Multi agent patterns are well-suited for tasks where feedback from multiple personas is beneficial in accomplishing the task. They are useful when parallelization across distinct tasks or workflows is required, allowing individual agents to proceed with their next steps without being hindered by the state of tasks handled by others.

Feedback and Human in the Loop

Language models tend to commit to an answer earlier in their response, which can cause a snowball effect of increasing diversion from their goal state [10]. By implementing feedback, agents are much more likely to correct their course and reach their goal. Human oversight improves the immediate outcome by aligning the agents responses more closely with human expectations, yielding more reliable and trustworthy results [11, 8]. Agents can be susceptible to feedback from other agents, even if the feedback is not sound. This can lead the agent team to generate a faulty plan which diverts them from their objective [12].

Information Sharing and Communication

Multi-agent patterns have a greater tendency to get caught up in niceties and ask one another things like how are you, while single agent patterns tend to stay focused on the task at hand since there is no team dynamic to manage. This can be mitigated by robust prompting. In vertical architectures, agents can fail to send critical information to their supporting agents not realizing the other agents arent privy to necessary information to complete their task. This failure can lead to confusion in the team or hallucination in the results. One approach to address this issue is to explicitly include information about access rights in the system prompt so that the agents have contextually appropriate interactions.

Impact of Role Definition and Dynamic Teams

Clear role definition is critical for both single and multi-agent architectures. Role definition ensures that the agents understands their assigned role, stay focused on the provided task, execute the proper tools, and minimizes hallucination of other capabilities. Establishing a clear group leader improves the overall performance of multi-agent teams by streamlining task assignment. Dynamic teams where agents are brought in and out of the system based on need have also been shown to be effective. This ensures that all agents participating in the tasks are strong contributors.

Summary of Key Insights

The key insights discussed suggest that the best agent architecture varies based on use case. Regardless of the architecture selected, the best performing agent systems tend to incorporate at least one of the following approaches: well defined system prompts, clear leadership and task division, dedicated reasoning / planning- execution evaluation phases, dynamic team structures, human or agentic feedback, and intelligent message filtering. Architectures that leverage these techniques are more effective across a variety of benchmarks and problem types.

Our meta-analysis aims to provide a holistic understanding of the current AI agent landscape and offer insight for those building with existing agent architectures or developing custom agent architectures. There are notable limitations and areas for future improvement in the design and development of autonomous AI agents such as a lack of comprehensive agent benchmarks, real world applicability, and the mitigation of harmful language model biases. These areas will need to be addressed in the near-term to enable reliable agents.

Here is the original post:

The Landscape of Emerging AI Agent Architectures for Reasoning, Planning, and Tool Calling: A - Towards Data Science

Invafresh names Andrew Cron Chief Data Officer – Supermarket News

Invafresh today announced Andrew Cron, Ph.D., has joined the company as chief data officer.

Cron is an accomplished executive-level data scientist and technology thought leader with a proven track record in driving transformational business results by bridging the gap between cutting-edge technology innovation and real-world application.

He joins Invafresh from 84.51, Krogers retail data science, insights, and media company, where he was SVP, chief scientist. In this role, directing a 100-member cross functional team comprising of researchers, data scientists, and engineers, Cronbuilt and led the R&D vision and strategy for Krogers data analytics and data monetization subsidiary, enhancing Krogers in-store and supply chain operations, as well as CPG partner product performance. Prior to 84.51, Cronheld senior data scientist roles with Citadel LLC, Weinrub Analytics and MaxPoint.

I am excited to join the Invafresh team as this is a pivotal time to be in grocery retail technology given the applicability AI has in addressing industry-wide challenges, said Cron. I look forward to partnering with our customers around the world to transform AI technology into tangible business solutions that will help optimize their fresh food operations.

AI technologies are revolutionizing the way grocery retailers operate, enabling them to streamline supply chains, enhance customer experiences, drive labor efficiencies, reduce food waste, and use technology to guide and accelerate decision making, said Tim Spencer, chief executive officer at Invafresh. Invafresh is at the forefront of accelerating the AI transformation of fresh food retail operations worldwide and I look forward to Andrews leadership in strengthening our capabilities in this area to deliver meaningful value to our customers.

Read the original here:

Invafresh names Andrew Cron Chief Data Officer - Supermarket News

Beyond Data Scientist: 10 Data Science Roles are Worth Applying For – Analytics Insight

In the ever-evolving landscape of data science, the demand for skilled professionals continues to soar. While the role of a data scientist is well-known, there exists a plethora of other specialized positions within the field, each offering unique opportunities for career growth and impact. In this article, we explore 10 data science roles beyond the traditional data scientist position that are worth considering and applying for in todays job market.

Data engineers play a crucial role in designing, building, and maintaining the infrastructure needed to support data-driven applications and analytics. They are responsible for managing large volumes of data, optimizing data pipelines, and ensuring data quality and reliability.

Machine learning engineers focus on developing and deploying machine learning models into production environments. They work closely with data scientists to turn research prototypes into scalable and robust machine learning systems that can make predictions and decisions autonomously.

Data analysts are responsible for interpreting and analyzing data to extract insights and inform decision-making. They work with stakeholders to understand business requirements, perform data analysis, and communicate findings through reports, dashboards, and visualizations.

BI developers design and build business intelligence solutions that enable organizations to gather, store, and analyze data to support strategic decision-making. They develop data models, design dashboards, and create interactive reports to provide actionable insights to business users.

Data architects design and implement the overall structure of an organizations data ecosystem. They define data architecture standards, design data models, and develop strategies for data integration, storage, and governance to ensure data consistency and accessibility across the organization.

Data product managers oversee the development and delivery of data-driven products and services. They work closely with cross-functional teams to define product requirements, prioritize features, and ensure alignment with business objectives and user needs.

In addition to the traditional data scientist role, there are specialized positions such as NLP (Natural Language Processing) Scientist, Computer Vision Scientist, and AI Research Scientist. These roles focus on applying advanced techniques and algorithms to solve specific problems in areas such as language understanding, image recognition, and artificial intelligence.

With increasing concerns about data privacy and regulatory compliance, organizations are hiring data privacy officers to ensure that data handling practices adhere to legal and ethical standards. Data privacy officers develop and implement privacy policies, conduct privacy impact assessments, and oversee compliance efforts.

Data governance managers are responsible for establishing and enforcing policies, procedures, and standards for managing data assets effectively. They work with stakeholders to define data governance frameworks, establish data quality metrics, and monitor compliance with data governance policies.

Data science consultants provide strategic advice and technical expertise to help organizations leverage data science and analytics to solve business challenges and achieve strategic goals. They work on a project basis, collaborating with clients to develop customized solutions and drive innovation through data-driven insights.

In conclusion, the field of data science offers a diverse array of career opportunities beyond the traditional data scientist role. Whether youre passionate about engineering data pipelines, building machine learning models, or driving strategic decision-making, there are plenty of exciting roles to explore within the data science domain. By considering these 10 data science roles and their respective responsibilities, you can identify the best fit for your skills, interests, and career aspirations in the dynamic world of data science.

Join our WhatsApp and Telegram Community to Get Regular Top Tech Updates

Read the original:

Beyond Data Scientist: 10 Data Science Roles are Worth Applying For - Analytics Insight

Avoiding abuse and misuse of T-test and ANOVA: Regression for categorical responses – Towards Data Science

We do the model comparison using the the loo package (9, 10) for leave-one-out cross validation. For an alternative approach using the WAIC criteria (11) I suggest you read this post also published by TDS Editors.

Under this scheme, the models have very similar performance. In fact, the first model is slightly better for out-of-sample predictions. Accounting for variance did not help much in this particular case, where (perhaps) relying on informative priors can unlock the next step of scientific inference.

I would appreciate your comments or feedback letting me know if this journey was useful to you. If you want more quality content on data science and other topics, you might consider becoming a medium member.

In the future, you may find an updated version of this post on my GitHub site.

1.M. Bieber, J. Gronewold, A.-C. Scharf, M. K. Schuhmann, F. Langhauser, S. Hopp, S. Mencl, E. Geuss, J. Leinweber, J. Guthmann, T. R. Doeppner, C. Kleinschnitz, G. Stoll, P. Kraft, D. M. Hermann, Validity and Reliability of Neurological Scores in Mice Exposed to Middle Cerebral Artery Occlusion. Stroke. 50, 28752882 (2019).

2. P.-C. Brkner, M. Vuorre, Ordinal Regression Models in Psychology: A Tutorial. Advances in Methods and Practices in Psychological Science. 2, 77101 (2019).

3. G. Gigerenzer, Mindless statistics. The Journal of Socio-Economics. 33, 587606 (2004).

4. P.-C. Brkner, Brms: An r package for bayesian multilevel models using stan. 80 (2017), doi:10.18637/jss.v080.i01.

5. H. Wickham, M. Averick, J. Bryan, W. Chang, L. D. McGowan, R. Franois, G. Grolemund, A. Hayes, L. Henry, J. Hester, M. Kuhn, T. L. Pedersen, E. Miller, S. M. Bache, K. Mller, J. Ooms, D. Robinson, D. P. Seidel, V. Spinu, K. Takahashi, D. Vaughan, C. Wilke, K. Woo, H. Yutani, Welcome to the tidyverse. 4, 1686 (2019).

6. D. Makowski, M. S. Ben-Shachar, D. Ldecke, bayestestR: Describing effects and their uncertainty, existence and significance within the bayesian framework. 4, 1541 (2019).

7. R. V. Lenth, Emmeans: Estimated marginal means, aka least-squares means (2023) (available at https://CRAN.R-project.org/package=emmeans).

8. R. McElreath, Statistical rethinking (Chapman; Hall/CRC, 2020; http://dx.doi.org/10.1201/9780429029608).

9. A. Vehtari, J. Gabry, M. Magnusson, Y. Yao, P.-C. Brkner, T. Paananen, A. Gelman, Loo: Efficient leave-one-out cross-validation and WAIC for bayesian models (2022) (available at https://mc-stan.org/loo/).

10. A. Vehtari, A. Gelman, J. Gabry, Practical Bayesian model evaluation using leave-one-out cross-validation and WAIC. Statistics and Computing. 27, 14131432 (2016).

11. A. Gelman, J. Hwang, A. Vehtari, Understanding predictive information criteria for Bayesian models. Statistics and Computing. 24, 9971016 (2013).

Read the original post:

Avoiding abuse and misuse of T-test and ANOVA: Regression for categorical responses - Towards Data Science

Researcher uses data science to address homelessness | UNC-Chapel Hill – The University of North Carolina at Chapel Hill

In the U.S., more than650,000 people dont have homes up 12% in 2023. Thats the largest jump seen since the government began collecting this data in 2007. The Triangle is no exception.More than 6,000 people identify as homelessin Raleigh and Wake County. Durham now hastwice as many unsheltered individualsas in 2020.

These numbers driveHsun-Ta Hsu, whos spent the last decade working with some of the largest homeless populations in the country, in Los Angeles and St. Louis, using innovative tools to address this problem.

In July 2023, Hsus unique skillset led him to Carolina, where he is a professor in both theUNC School of Social Workand the newUNC School of Data Science and Society.

Dr. Hsu is a prime example of how interdisciplinary data science can create insights that transform a seemingly intractable, multilevel social issue into something solvable, SDSS DeanStan Ahaltsays.

Ramona Denby-Brinson, dean of social work, agrees about Hsus skills. His work advances our understanding of neighborhood structures, the development of effective intervention programs and services, and how we can employ social networks in more practical terms to produce better health and behavioral outcomes for the unhoused.

A human right

Hsu learned about social work in high school when his adviser recommended that he major in it based on his background and interests. In college, he earned bachelors, masters and doctoral degrees in social work.

I had relatives and people I was close to who were suffering with mental health-related issues, including suicide attempts and substance abuse, he shares. When I was younger, I didnt know how to deal with it. So I was really thinking about that and I wanted to do something about it.

In 2010, at the start of his doctoral program at the University of Southern California, Hsu got his first look at the 50-block area of Los Angeles known as Skid Row. I saw a young mother in a wheelchair breastfeeding her baby, surrounded by tents, bad smells and extreme poverty, Hsu recalls. Thats not OK. To me, housing is a human right.

Hsu analyzed data from interviews with people housed by the Los Angeles Homeless Service Authority. He documented neighborhood characteristics for 50 blocks, a time-consuming, labor-intensive process that he thought technology could improve.

After a summer fellowship at theUSC Center for Artificial Intelligence in Society, he started developing a mapping tool that uses machine learning to automate the identification of objects like garbage and broken-down cars.

Community-centered research

Since 2010, cities across the country have used another tool, thevulnerability index, to prioritize who gets housing.

Its a triage tool like we use in the emergency room, Hsu explains. We are measuring how vulnerable one is on the street and then bumping them up on the priority list to get them housing.

In 2019, Hsu teamed up with CAIS researcher Eric Rice to improve this tool by combining demographic data with feedback from community stakeholders. They said they want to be considered for housing based on assets, not deficits. This super important feedback helped Hsu and Rice revise the vulnerability index survey to include questions focused on an individuals positive traits.

Now Hsu is bringing this project model to rural communities, where nearly87,000 Americans experience homelessness. Hsu believes his research in both rural and urban homeless populations will aid future projects in North Carolina and beyond.

Homelessness is a national issue, Ahalt stresses. This research will create a replicable process that can be used in North Carolina and across the country.

Read more about Hsun-Ta Hsus work.

The rest is here:

Researcher uses data science to address homelessness | UNC-Chapel Hill - The University of North Carolina at Chapel Hill

Using AI and Data Science to ReNOURISH Food Deserts – University of California San Diego

Twenty-four million Americans live in food deserts where ultraprocessed foods are abundant and fresh food is scarce, giving rise to large health disparities in diabetes and related cardiometabolic diseases. To address this problem, an interdisciplinary team of researchers from UC San Francisco and UC San Diego conceptualized the NOURISH platform, winning support last year from the U.S. National Science Foundation (NSF)Convergence Accelerator program to design the tool. Now, with continued NSF and U.S. Department of Agriculture funding, the team of experts has moved into the platform-building phase.

NOURISH is meant to provide small business owners in food desert communities with access to loans and grants, online maps that optimize the placement of fresh food outlets for foot traffic, help with navigating the convoluted business permitting process and AI-enabled guidance on affordable ways to locally source fresh ingredients.

Our solution complements government efforts to get fresh food into food deserts by incentivizing grocery stores and big box outlets to sell more fresh food, said Laura Schmidt of UC San Francisco, principal investigator for the project. But our approach builds upon the often-overlooked assets of these communities, including the entrepreneurial talent of small business owners, rich and diverse food heritages, and an unmet demand for fresh food.

Under the leadership of Amarnath Gupta, a team of computer scientists, software developers and students at the San Diego Supercomputer Center (SDSC) at UC San Diego are combining government, private sector and crowdsourced information to create dynamic, interactive maps of local food systems across the U.S. Gupta is a leading computer scientist in the Cyberinfrastructure and Convergence Research and Education (CICORE) Division at SDSC, directed by Ilkay Altintas.

NOURISH embodies our vision at SDSCs CICORE Division, where our deep expertise in data science and knowledge management seamlessly integrates with the diverse needs of our interdisciplinary and cross-sector partners. Together, we co-create solutions that are not just equitable but deeply impactful, tackling complex societal challenges head-on, said Altintas, who also serves as SCSCs chief data science officer. In an era where access to equitable access to fresh food is still hard, NOURISH emerges as a solution to leverage cutting-edge technology to bridge the gap between communities and an ecosystem of entrepreneurship, innovation and cultural diversity. We look forward to seeing the growing impact of this project over the years to come.

Accessible from a mobile phone in multiple languages, the NOURISH platform will include patented recommendation algorithms that customize business plans based on local consumer preferences for price, convenience and flavor.

Recent advances in scalable data systems and artificial intelligence give us an unprecedented opportunity to use NOURISH to democratize data access, creating a more level playing field between large food companies and small businesses, Gupta said.

Small businesses have relatively low start-up costs, are adaptive to local needs and can help to keep economic resources circulating within low-income communities. Community partners assisting with NOURISH also emphasize the benefits of promoting culturally appropriate food.

A major asset of so-called food deserts are immigrants who bring diverse cuisines featuring traditional dishes that are typically healthier than the standard American diet. This platform will help people from the community make wholesome food for the community, said Paul Watson, a California-based food equity advocate and director of community engagement for NOURISH.

Other scientists on the team include Keith Pezzoli and Ilya Zaslavsky (UC San Diego), Hans Taparia (New York University), Tera Fazzino (University of Kansas) and Matthew Lange (IC-FOODS). In 2024-25, the NOURISH team will test the platform in lower-income areas within San Diego and Imperial counties in California, and then scale it nationally.

Originally posted here:

Using AI and Data Science to ReNOURISH Food Deserts - University of California San Diego

From Economics to Electrocardiograms, Data Science Projects Get a Boost From New Seed Grants – University of Utah Health Sciences

Explaining Data Evolution Anna Fariha, PhD (School of Computing) Nina de Lacy, MD (Department of Psychiatry)

Scalable and Information-Rich Sequence Search Over SRA for Advanced Biological Analyses Prashant Pandey, PhD (School of Computing) Aaron Quinlan, PhD (Departments of Human Genetics and Biomedical Informatics)

Connecting the Metabolite-Protein Interactome: Precision Diet and Drug Synergy for Enhanced Cancer Care Mary Playdon, PhD (Departments of Nutrition & Integrative Physiology and Population Health Sciences) Kevin Hicks, PhD (Department of Biochemistry) Aik Choon Tan, PhD (Departments of Oncology and Biomedical Informatics)

Information Theoretic Approaches to Causal Inference Ellis Scharfenaker, PhD (Department of Economics) Braxton Osting, PhD (Department of Mathematics)

Automated Live Meta-Analysis of Clinical Outcomes Using Generative AI Fatemeh Shah-Mohammadi, PhD (Department of Biomedical Informatics) Joseph Finkelstein, MD, PhD (Department of Biomedical Informatics)

Modeling the Effect of Artificial Nature Exposure on Brain Health in Bed-bound Populations Using Variational Autoencoders Elliot Smith, PhD (Department of Neurosurgery) Jeanine Stefanucci, PhD (Department of Psychology)

Using Controlled Animal ECG Recordings for Machine Learning-Based Prediction of Myocardial Ischemia Outcomes Tolga Tasdizen, PhD (Department of Electrical & Computer Engineering and School of Computing) Ben Steinberg, MD (Division of Cardiovascular Medicine) Rob MacLeod, PhD (Departments of Biomedical Engineering and Internal Medicine)

See the article here:

From Economics to Electrocardiograms, Data Science Projects Get a Boost From New Seed Grants - University of Utah Health Sciences

How to Empower Pandas with GPUs. A quick introduction to cuDF, an NVIDIA | by Naser Tamimi | Apr, 2024 – Towards Data Science

DATA SCIENCE A quick introduction to cuDF, an NVIDIA framework for accelerating Pandas 6 min read

Pandas remains a crucial tool in data analytics and machine learning endeavors, offering extensive capabilities for tasks such as data reading, transformation, cleaning, and writing. However, its efficiency with large datasets is somewhat limited, hindering its application in production environments or for constructing resilient data pipelines, despite its widespread use in data science projects.

Similar to Apache Spark, Pandas loads the data into memory for computation and transformation. But unlike Spark, Pandas is not a a distributed compute platform, and therefore everything must be done on a single system CPU and memory (single-node processing). This feature limits the use of Pandas in two ways:

The first issue is addressed by frameworks such as Dask. Dask DataFrame helps you process large tabular data by parallelizing Pandas on a distributed cluster of computers. In many ways, Pandas empowered by Dask is similar to Apache Spark (however, still Spark can handle large datasets more efficiently and thats why it is a preffered tool among data engineers).

Although Dask enables parallel processing of large datasets across a cluster of machines, in reality, the data for most machine learning projects can be accommodated within a single systems memory. Consequently, employing a cluster of machines for such projects might be excessive. Thus, there is a need for a tool that efficiently executes Pandas operations in parallel on a single machine, addressing the second issue mentioned earlier.

Whenever someone talks about parallel processing, the first word that comes to most engineers' minds is GPU. For a long time, it was a wish to run Pandas on GPU for efficient parallel computing. The wish came true with the introduction of NVIDIA RAPIDS cuDF. cuDF (pronounced KOO-dee-eff) is a GPU DataFrame library for

Read the original here:

How to Empower Pandas with GPUs. A quick introduction to cuDF, an NVIDIA | by Naser Tamimi | Apr, 2024 - Towards Data Science