Category Archives: Data Science
Top 10 Best Machine Learning Companies to Join in 2022 – Analytics Insight
Machine learning is a blessing. Here are the 10 best machine learning companies to join in 2022
Machine learning is a blessing. The industrial sector saw a radical shift when machine learning and AI came into the limelight. Machine learning companies are gradually evolving at a faster pace and have emerged as one of the key players of IT firms. Machine learning refers to the development of intelligent algorithms and statistical modelling, that aids in improving programming, without coding them explicitly. For example, ML can make a predictive analysis app more precise, with the passing time. ML frameworks and models require an amalgamation of data science, engineering, and development skills. As we are becoming completely dependent on technology for making our lives faster and smoother, machine learning has also become an integral part of our lives. It is now widely accessed by various organizations on this planet. They have started building in-house data science teams. Some of these teams primarily focus on analysing business data, to generate valuable insights and the rest try to incorporate machine learning capabilities into their companys products.
Share This ArticleDo the sharing thingy
About AuthorMore info about author
More:
Top 10 Best Machine Learning Companies to Join in 2022 - Analytics Insight
A Guide to ECCO: Python Based Tool for Explainability of Transformers – Analytics India Magazine
Accountability is required for any decision-making tool in an organization. Machine learning models are already being used to automate time-consuming administrative tasks and to make complex business decisions. To ensure proper security of the model and business decisions, scientists and engineers must understand the inner mechanics of their models, which is commonly referred to as a black box. This is no longer the case, as various tools, such as ELI5, are available to track the inner mechanics of the model. In this article, well look at how to explain the inner workings of language models like transformers using a toolbox called ECCO. The main points to be covered in this article are listed below.
Lets start the discussion by understanding the explainability of machine learning models.
Explainability in machine learning refers to the process of explaining a machine learning models decision to a human. The term model explainability refers to the ability of a human to understand an algorithms decision or output. Its the process of deciphering the reasoning behind a machine learning models decisions and outcomes. With black box machine learning models, which develop and learn directly from data without human supervision or guidance, this is an important concept to understand.
A human developer would traditionally write the code for a system or model. The system evolves from the data with machine learning. Machine learning will be used to improve the algorithms ability to perform a specific task or action by learning from data. Because the underlying functionality of the machine learning model was developed by the system itself, it can be difficult to understand why the system made a particular decision once it is deployed.
Machine learning models are used to classify new data or predict trends by learning relationships between input and output data. The model will identify these patterns and relationships within the dataset. This means that the deployed model will make decisions based on patterns and relationships that human developers may not be aware of. The explainability process aids human specialists in comprehending the decisions algorithm. After that, the model can be explained to non-technical stakeholders.
Machine learning explainability can be achieved using a variety of tools and techniques that vary in approach and machine learning model type. Traditional machine learning models may be simpler to comprehend and explain, but more complex models, such as deep neural networks, can be extremely difficult to grasp.
When machine learning has a negative impact on business profits, it earns a bad reputation. This is frequently the result of a misalignment between the data science and business teams. There are a few areas where Explainability heals based on this, such as,
Understanding how your models make decisions reveals previously unknown vulnerabilities and flaws. Control is simple with these insights. When applied across all models in production, the ability to quickly identify and correct mistakes in low-risk situations adds up.
In high-risk industries like healthcare and finance, trust is critical. Before ML solutions can be used and trusted, all stakeholders must have a thorough understanding of what the model does. If you claim that your model is better at making decisions and detecting patterns than humans, you must be able to back it up with evidence. Experts in the field are understandably skeptical of any technology that claims to be able to see more than they can.
When a model makes a bad or rogue decision, its critical to understand the factors that led to that decision, as well as who is to blame for the failure, in order to avoid similar issues in the future. Data science teams can use explainability to give organizations more control over AI tools.
The terms explainability and interpretability are frequently used interchangeably in the disciplines of machine learning and artificial intelligence. While they are very similar, it is instructive to note the distinctions, if only to get a sense of how tough things may become as you advance deeper into machine learning systems.
The degree to which a cause and effect may be observed inside a system is referred to as interpretability. To put it another way, its your capacity to predict what will happen if the input or computational parameters are changed.
Explainability, on the other hand, relates to how well a machines or deep learning systems internal mechanics can be articulated in human terms. Its easy to ignore the subtle contrast between interpretability and comprehension, but consider this: interpretability is the ability to comprehend mechanics without necessarily knowing why. The ability to explain what is happening in depth is referred to as explainability.
Many recent advances in NLP have been powered by the transformer architecture, and until now, we had no idea why Transformer-based NLP models have been so successful in recent years. To improve the transparency of Transformer-based language models, ECCO an open-source library for the explainability of Transformer-based NLP models was created.
ECCO offers tools and interactive explorable explanations to help with the examination and intuition of terms, such as Input Saliency, which visualizes the token importance for a given sentence. Hidden State Evaluation is applied to all layers of a model to determine the role of each layer. Non-negative matrix factorization of neuron activations was used to uncover underlying patterns of neuron firings, revealing firing patterns of linguistic properties of input tokens, and neuron activation tell us how a group of neurons spikes or responds while making a prediction.
Now in this section, we will take a look at how ECCO can be used to understand the working of various transformers models while predicting the sequence-based output. Majorly well see how weights are distributed at the final layer while predicting the next sequence and will also analyze all layers of the selected model.
To start with ECCO we can install it using the pip command as! pip install Ecco
And also make sure you have also installed the Pytorch.
First, we will start with generating a single token by passing a random string to the model. The GPT2 is used due to its superiority for generating the next sequence as a human does. The below code shows how we load the pre-trained model and how to use it for the prediction. The below generate method takes an input sequence and additionally there we can pass how many tokens we need to generate from the model by specifying generate= some number.
While initializing the pre-trained model we set activation=True that we capture all the firing status of the neurons.
Now well generate a token using the generate method.
From the method, token 6 and 5 is generated as first and of respectively.
The model has a total of 6 decoder layers and the last layer is the decision layer where the appropriate token is chosen.
Now we will observe the status of the last layers and see what are the top 15 tokens that the model has considered. Here we observe the status for position / token 6 and this can be achieved by output.layer_predictions method as below.
output.layer_predictions(position=6, layer=5, topk=15)
As we can see, the token first comes up with a higher contribution.
Similarly, we can check how different tokens would perform at the output layer. This can be done by explicitly passing the token numbers inside the method ranking_watch. However, tokens can be easily generated by using the pre-trained model that we have selected initially.
Below are the generated token IDs.
Now well supply these IDs to see the rankings.
output.rankings_watch(watch=[262, 717, 621], position=6)
At the decision layer, we can see the first rank is achieved by token first and the rest not even closer to it. Thus we can say the model has correctly identified the next token and did assign proper weights for possible tokens.
We have seen what the explainability of a Model is and how important it is when it comes to deploying such a model to the production level in this article. Tools are needed to aid debugging models, explain their behavior, and develop intuitions about their inner mechanics as language models become more common. Ecco is one such tool that combines ease of use, visual interactive explorable, and a variety of model explainability methods. This article focused on the ML models explainability and a glimpse of the ECCO Toolbox.
Read this article:
A Guide to ECCO: Python Based Tool for Explainability of Transformers - Analytics India Magazine
What to know about the Minnesota redistricting plans going before a special judicial panel this week – MinnPost
A passel of lawyers will gather Tuesday morning in a large conference room in the Minnesota Judicial Center and take their last shot at influencing the special five-judge panel charged with drawing new congressional and legislative districts for the state.
The oral arguments in Wattson v. Simon will be the final public part of the process triggered 10 months ago by a lawsuit asking the Minnesota Supreme Court to address the assertion that the 2020 Census made the states current political lines unconstitutional.
After the hearing, the panel will spend six weeks doing two things. The first will be drawing eight new congressional districts, 67 state Senate districts and 134 state House districts. The second will be waiting, at least until Feb. 15, to be certain that the divided state Legislature will fail in its redistricting duties.
Lawyers for each of the four groups proposing new maps, known as intervenors, will be given time to make the case that their vision is the correct one and that the visions of the other three plaintiffs are not.
Article continues after advertisement
On Dec. 7, the four groups filed documents detailing their plans; on Dec. 17, all four filed briefs defending their plan and critiquing the plans of the others. Those briefs give an early look at what will be talked about during oral arguments Tuesday.
A typical civil case involves two parties: a plaintiff and a respondent. This one involves five: four that have proposed maps and Secretary of State Steve Simon, who was sued. The map-makers are known as the Wattson Plaintiffs (for lead plaintiff Peter Wattson, a former legislative lawyer involved in past redistricting efforts); the Anderson Plaintiffs (representing Republican Party interests); the Sachs Plaintiffs (representing DFL interests); and the Corrie Plaintiffs (for lead plaintiff Bruce Corrie, who are advocating for maximum representation for communities of color).
A fifth group, though not formal intervenors, has submitted a series of friend of court filings with a December 8 filing accepted by the court but a Dec. 29 filing commenting on the four intervenors plans rejected [December 8 Minnesota Special Redistricting Panel Brief, December 29 Minnesota Motion for Leave]. Calling themselves the Citizen Data Scientists, the group of 12 Minnesota residents is made up of professors, practitioners, and researchers in data science, computer science, mathematics, statistics, and engineering at some of Minnesotas leading institutions of higher education, who applied computational redistricting, a relatively new field that uses high-performance computers and optimization algorithms to systematically search through millions of possible combinations of district boundaries.
Here is a sampling of how the four intervenors defended their own work and attacked the others in the lengthy briefs that were filed in mid-December.
The Wattson plaintiffs proposal follows a least-change approach that advocates that court-drawn lines make just enough changes to restore population balance while following other legal mandates set by the panel. Based on the 2020 Census, Minnesotas congressional districts should have 713,312 residents, state Senate districts should have 85,172 and state House districts should have 42,586.
The other principles set out by the judicial panel include: not harming communities of color; not overly dividing local government boundaries; not dividing the reservations of American Indian tribes; crafting districts that are contiguous and convenient for voters; preserving communities of people with shared interests; and avoiding drawing lines with the purpose of protecting, promoting or defeating any incumbent, candidate or political party.
The Wattson plaintiffs proposal for new congressional districts.
The plans submitted by the other parties in this matter fail to adhere to this Panels redistricting principles for some obvious reasons, and some not so obvious reasons A less obvious but very important reason is that the plans of the Anderson Plaintiffs and Sachs Plaintiffs were drawn for the purpose of promoting, protecting or defeating an incumbent, candidate or party, notes the Wattson plaintiffs brief. The districts created by these parties can be explained on no ground other than attempting to gain a partisan advantage.
Article continues after advertisement
The Wattson Plaintiffs have argued that the only way to know if a plan was drawn to help an incumbent or party is to know where incumbents live and how proposed lines would impact future elections. This comes despite the panels assertion that it will not draw districts based on the residence of incumbent office holders and will not consider past election results when drawing districts.
Wattson forges ahead anyway, citing a partisan index the plaintiffs created to apply past election results to new lines.
One example that Wattson cites is how the DFL-friendly Sachs Plaintiffs plan shifts voters from the 3rd Congressional District (now held by DFL Rep. Dean Phillips) and the 5th Congressional District (now held by DFL Rep. Ilhan Omar) to make the 2nd Congressional District (now held by DFL Rep. Angie Craig) safer for Democrats.
The net effect of these changes is that CD 5 is much less convenient. It is sandwiched between CD 3 and CD 4 and is shaped like a T or a hammer, the Wattson brief states.
The Wattson brief also points out that the Corrie Plaintiffs new 8th Congressional District includes three GOP incumbents: U.S. Reps. Pete Stauber, Michelle Fischbach and Tom Emmer, while the DFL-leaning Sachs Plaintiffs plan puts both Emmer and Fischbach in the same district.
By just narrowly including Representative Emmer in CD 7 (Corrie Plaintiffs Plan) and narrowly including Representative Fischbach in CD 6 (Sachs Plaintiffs Plan), with no justification other than population, it is apparent that these pairings were done to defeat Republican incumbents, the Wattson brief states.
The GOP-leaning group of intervenors said they base their congressional plan on a geographic distribution of seats established in previous redistricting processes.
Each of the Opposing Parties congressional redistricting plans propose drastic reconfigurations to Minnesotas existing congressional districts and fail to meet this Panels redistricting criteria, the Anderson brief states, by combining rural and suburban communities into the same district. Doing so negatively impacts the ability for rural voters to elect representatives that reflect their priorities and concerns.
Article continues after advertisement
The Anderson Congressional Plan, on the other hand preserves the unique interests of rural, suburban/exurban, and urban Minnesotans.
Anderson takes issue with a new 8th Congressional District proposed by the Corrie Plaintiffs that reaches across the northern part of the state from North Dakota to Lake Superior.
The Anderson plaintiffs proposal for new congressional districts.
Anderson also accuses DFL-leaning plans of helping the DFL win more seats in Congress: By moving first ring suburbs, which have natural affinities with and similarities to Minneapolis and St. Paul, to districts comprised largely of highly suburban and exurban areas, these parties put more DFL-leaning voters in the perennially toss-up Third and Second districts, Anderson wrote. At the same time, removing first ring suburbs and adding outer suburban voters to the urban Fourth and Fifth districts pose no real risk to DFL candidates, incumbents, or the party, because the Fourth and Fifth districts have had highly reliable DFL majorities for decades.
The DFL-leaning group relies heavily on testimony given during the five-judge panels public hearings in October and criticizes others especially Anderson and Wattson for not taking that testimony into account. (For their part, those intervenors say Sachs cherry picks testimony that supports their decisions and disregards others.)
Sachs also accuses the Wattson plaintiffs of overly strict adherence to its least-change philosophy. Rather than draw districts that are responsive to the states geography and demographics, they instead pursue what they characterize as a least-change approach, one that rigidly focuses on calcified lines on a map and not the wishes and needs of Minnesotans statewide, the Sachs brief states. Their overemphasis on staticity for its own sake has produced proposed maps that are non-responsive to the clear wishes of Minnesotans as expressed to the Panel and that will consequently fail to accurately reflect the human geography of the state.
The Sachs plaintiffs proposal for new congressional districts.
Sachs also criticizes Wattson for using election analyses and incumbent location data. The Sachs Plaintiffs maintain that these sorts of partisan considerations ask the Panel to delve into troubling political waters, Sachs stated. Whether the parties proposed plans avoid impermissible political entanglements should instead be judged based on the degree to which they otherwise satisfy the Panels neutral redistricting criteria, particularly evidence in the record regarding the suitability of joining communities within the same district and dividing others among different districts.
Article continues after advertisement
Sachs also objects that Anderson and Wattson continue to have a First Congressional District that runs across the entire border with Iowa, accusing them of slavish devotion to prior district lines. The Sachs plan instead joins the southwest counties with a new 7th Congressional district that would run north and south from Iowa to Canada.
While both Corrie and Sachs criticize the Wattson plan for the least-change approach and a desire to avoid splitting local governments and precincts, they do so with different conclusions. Said Sachs: the Wattson Plaintiffs have ignored the Redistricting Principles laid out by this Panel, and instead prioritized their own principles, particularly preserving voting precincts and ensuring political competitiveness based on past election results.
But Corrie sees much different motives. In stark contrast to the Panels directive, the Wattson brief makes clear that its maps were created to ensure each incumbent is protected and unabashedly describes how districts were created based on where incumbents live and how to solidify their votes. Throughout their discussion, the Wattson Plaintiffs make scant mention of Minnesotas BIPOC communities. Rather, they pursue incumbent protection in the guise of protecting minority voting rights, perhaps hoping this Panel will not see they have directly contravened this Panels Redistricting Principles.
The Corrie plaintiffs proposal for new congressional districts.
The Corrie Plaintiffs House Plan has 24 districts with 30% or greater minority voting-age population. The Sachs Plaintiffs House Plan also has 24, but the Wattson Plaintiffs has only 21, and the Anderson Plaintiffs has only 18. The Corrie House Plan is the only plan that creates a district (HD 2B) where American Indian/Native American residents constitute 44.5% of the district population, giving this community the ability to elect candidates of choice when voting in alliance with others.
And Corrie explains its choice to spread the 8th Congressional District from east to west as a way to get the states tribal nations into a single district.
As the only map proposal that places all of northern Minnesota in one district, thereby bringing together the three largest American Indian reservations (Red Lake Nation, White Earth Nation, and Leech Lake Band of Ojibwe) as well as four other tribal reservations (such as Bois Forte Band of Chippewa, Fond du Lac Band of Lake Superior Chippewa, and Mille Lacs Band of Ojibwe, Grand Portage Band of Lake Superior Chippewa) and trust lands, the Corrie Congressional Map is the only map that abides by the Courts Redistricting Principles.
Here is the original post:
Upcoming NSTDA Supercomputer in Thailand to Use Nvidia A100 GPUs – CDOTrends
Thailands National Science and Technology Development Agency (NSTDA) upcoming supercomputer will harness hundreds of GPUs, making it the largest public high-performance computing system in Southeast Asia, says Nvidia.
Powered by 704 Nvidia A100 Tensor Core GPUs, the new system will be 30 times faster than the current TARA HPC system. According to information from Nvidias product page, the A100 is available in 40GB or 80GB variants and offers up to 294 times higher AI inference performance over traditional CPUs.
The new supercomputer will be hosted at the NSTDA Supercomputer Centre (ThaiSC) to drive research by engineers and computational and data scientists from academia, government, and industry sectors. It is expected to support research projects in areas such as pharmaceuticals, renewable energy, and weather forecasting.
The new supercomputer at NSTDA will expand and enhance research in Thailand, speeding up the development of breakthroughs that benefit individuals and industries in the country, said Dennis Ang the senior director of enterprise business for worldwide field operations in the SEA and ANZ region at Nvidia.
NVIDIA A100 incorporates building blocks across hardware, networking, software, libraries, optimized AI models, and applications to enable extreme performance for AI and HPC, said Ang.
We chose NVIDIA A100 because it is currently the leading solution for HPC-AI in the market. Even more important is that many HPC-AI software applications are well supported by NVIDIA technology, and the list will keep growing, explained Manaschai Kunaseth, chief of operations at ThaiSC.
When operational, the additional power of the new supercomputer will allow users at ThaiSC to scale up existing research projects. Specifically, the new supercomputer will accelerate innovation for Thailands efforts with more advanced modeling, simulation, AI, and analytics capabilities.
Kwanchiva Thangthai, of the National Electronics and Computer Technology Centers Speech and Text Understanding Team, expects to see massive efficiency gains in speech recognition research pipelines. We can gain competitive performance and provide a free-of-charge Thai speech-to-text service for everyone via AIForThai, she said.
The supercomputer is expected to commence operation in the second half of 2022.
Image credit: iStockphoto/sdecoret
Original post:
Upcoming NSTDA Supercomputer in Thailand to Use Nvidia A100 GPUs - CDOTrends
January 2022: Insight into how metabolites affect health aided by new data platforms – Environmental Factor Newsletter
Gary Siuzdak, Ph.D., from the Scripps Research Institute, highlighted exciting technologies that he said will advance the field of metabolomics and a wide range of scientific discovery, during a Dec. 7 NIEHS lecture. Metabolomics is the large-scale study of chemical reactions involving metabolites, which are small molecules that play important roles in cells, tissues, and organisms.
According to Siuzdak, research in this field originally focused on identifying metabolites that serve as biological signs of disease, which scientists call biomarkers. However, metabolomics has evolved into a more comprehensive tool for understanding how metabolites themselves can influence health and illness.
The most important area where metabolomics can be applied is in looking for active metabolites that affect physiology, Siuzdak said. For example, metabolites can impact and even improve the way we respond to medicine or exposure to toxic agents.
Siuzdak developed data analysis platforms called XCMS and METLIN that enable scientists to discover how metabolites can alter critical biological processes, and the tools have been cited in more than 10,000 scientific projects, he noted.
Through XCMS and METLIN, which now contains detailed data on 860,000 molecular standards, the Scripps Center for Metabolomics has strengthened research worldwide, across a variety of disciplines, said Siuzdak, the centers director.
Continued development of databases like METLIN is vital to success of the metabolomics field, noted David Crizer, Ph.D., a chemist in the NIEHS Division of the National Toxicology Program. He is a member of the institutes Metabolomics Cross-Divisional Group, which hosted Siuzdaks talk (see sidebar).
METLIN is designed to help scientists identify molecules in organisms, whether metabolites, toxicological agents, or other chemical entities, according to Siuzdak. He noted that the database encompasses more than 350 chemical classes, and there now are more than 50,000 registered users in 132 countries.
Our goal is to identify as many metabolites and other chemical entities as possible, and given the advances in other fields of biology, this data is long overdue, Siuzdak said.
We are finding metabolites that were previously unknown, quite regularly, he added. The more comprehensive METLIN is, the better chance we have of eventually identifying all molecules. To this end, I am constantly looking for ways to facilitate growth of the platform.
A metabolite called indole-3-propionic acid (IPA) is of particular interest to Siuzdak. IPA is a human gutderived metabolite originally identified by his lab in a 2009 paper in the Proceedings of the National Academy of Sciences, and it has since been examined in thousands of studies. Researchers have discovered that it is a multifunctional molecule that can aid immune function, among other roles.
In retrospect, it makes sense that a metabolite derived from a gut microbe could modulate the immune system, which is probably why it still generates so much excitement, he said.
IPA could be especially relevant with respect to autoimmune diseases, Siuzdak added.
For example, most people who die from COVID-19 dont succumb to the virus but from an overactive immune response that causes them to develop respiratory ailments, he said. A metabolite that modulates this effect could be very beneficial, noted Siuzdak.
Overall, we are pursuing one primary goal in the development of METLIN, which is to use experimental data generated from molecular standards to help identify these key, physiologically relevant molecules, he said.
Citation: Wikoff WR, Anfora AT, Liu J, Schultz PG, Lesley SA, Peters EC, Siuzdak G. 2009. Metabolomics analysis reveals large effects of gut microflora on mammalian blood metabolites. Proc Natl Acad Sci U S A 106(10):36983703.
(John Yewell is a contract writer for the NIEHS Office of Communications and Public Liaison.)
Read the original post:
Texas public universities to partner for data science camp – The Cougar – The Daily Cougar
By Susana Arreola December 16, 2021
Five universities will partner for a five-week data science camp. | File photo
With a high demand for a data science workforce that has a broad understanding of conventional energy, five major Texas public universities, including UH,are setting up a path to the industry.
The University will be leading a project in collaboration with UH-Downtown, UH-Victoria, UH-Clear Lake and Sam Houston State University. Next summer, all five schools, and several energy industry partners will train students in a five-week data science camp aimed at transitioning to cleaner energy.
Students will get useful training on basics of data science as well as topics on geoscience and public policy, all under the theme of energy and energy transition, said principal investigator for the program Mikyoung Jun. The goal here is to prepare them to be a future workforce for the industry and community.
Jun and 11 other faculty members fromthe five universities will serve as senior personnel. The projects faculty affiliates will also teach and mentor student participants.
Energy industry partners for the project include companies like ConocoPhillips, Schlumberger, Fugro, Quantico Energy Solutions, Shell and Xecta Web Technologies.
We have a website to inform people about our program as well as to take applications, Jun said. But we plan to utilize social media and other available channels to advertise. We have 40 spots available for each year, and so there may be some selection process depending on how many applications we get.
The program starts with a five-week summer boot camp. After, students will take more advanced courses on some of the topics in the fall at their home institution. They would then participate in team research projects with real data provided by the industry partners in the spring. The end goal is for students to get a summer internship the following summer, concluding the program.
The program is open to any major and no prerequisites are required. Both undergraduate and graduate students can apply.
Any student interested in learning about data science and how data science techniques can be applied to various energy-related problems would be welcome, Jun said.
[emailprotected]
Tags: data science, energy, research
Read more here:
Texas public universities to partner for data science camp - The Cougar - The Daily Cougar
Lecturer/Associate Professor in Computational Environmental Science and Data Analytics job with UNIVERSITY OF SOUTHAMPTON | 275934 – Times Higher…
School of Ocean and Earth Science
Location: National Oceanography Centre SouthamptonSalary: 39,739 to 65,107 - per annumFull Time PermanentClosing Date: Monday 31 January 2022Interview Date: To be confirmedReference: 1648921HN
Join us and be part of the future of Ocean and Earth science.
We are seeking to make a Faculty appointment in the area of Ocean and Earth as part of a series of 4 new appointments to further enhance our research-led education and extend the international reach of our research and enterprise activities. This will be the first part of a series planned appointments over the next five years. You will have an exciting opportunity to influence the Ocean and Earth sciences agenda within a world leading school at the University of Southampton. Our School takes pride in research that spans marine biology and ecology, physical oceanography, palaeoceanography and paleoclimate, marine biogeochemistry, geology and geophysics and geochemistry. We lead the prestigiousNERC INSPIRE Doctoral Training Programme (DTP) and are home to one of the largest graduate schools in Ocean and Earth Science in the world, with >150 PhD students. We are integral to both the interdisciplinary Southampton Marine and Maritime Institute and Institute for Life Sciences. We are home to the only Regius Chair in Ocean Sciences in the country. Based at the National Oceanography Centre Southampton site we have a range of unique and world class facilities. We collaborate broadly both nationally and internationally, including our close collaborative relationships with the independent National Oceanography Centre.
We seek applications from individuals using innovative computational approaches to pursue outstanding research questions at the interfaces among earth, ocean, climate and life sciences. You will have a growing international profile that adds a new dimension to the thriving research programs across our interdisciplinary School of Ocean and Earth Science (SOES). Based at the National Oceanography Centre Southampton, home to the Marine Autonomous Robotic Systems fleet, and with strong links to the Southampton Marine and Maritime Institute, GeoData institute, and world-class supercomputer facilities across the University, SOES is ideally located to support innovative research agendas in this field.
You will take a data and/or computer science approach to exploit the opportunities presented by vast data sets. For example datasets generated by increased deployment of autonomous systems featuring high-resolution sensor and imaging capability, or large equipment arrays. You might leverage those data to answer outstanding research questions spanning geospatial, climatic, geohazards, biological tracking or oceanographic fields. Your specialisms in data analytics might include the analysis of big data through Machine Learning, Artificial Intelligence, uncertainty quantification, and/or data assimilation and integration.
You will demonstrate a passion for education and an ability to enhance our undergraduate and postgraduate curriculum in Ocean and Earth science.
Strong applicants should demonstrate:
These posts are offered on a full-time, permanent basis.
We welcome applications from all candidates with an interest in the role, and those who are committed to helping us create an inclusive work environment. We encourage applications from candidates from Black, Asian and Minority Ethnic communities, people who identify as LGBTQ+ and people with disabilities.
Interviews will take place in person or online in February 2022. In the event that interviews are online, applicants who are made an offer will be invited to campus for further discussion.
What we can offer you
We are an ambitious, mutually supportive academic community and passionate about creating an environment where everyone can achieve their full potential in education, research and enterprise. As such, we hold both a departmental Athena SWAN bronze award. We value diversity and equality and recognise that employees may wish to have working patterns that fit with their caring responsibilities or work-life balance. Due consideration will be given to applicants who have had career breaks for reasons including maternity, paternity or adoption leave, disability or illness. We offer a generous holiday allowance and additional University closure days, subsidised health and fitness facilities and access to the Universities Superannuation Scheme (USS).
For further details about the post you are invited to contact the Head of School, Professor Mark Moore cmm1g06@soton.ac.uk 44 (0) 2380599006
This vacancy is for Lecturer / Associate Professor, please see the job descriptions showing the different levels of responsibilities and state in your application which Level you wish to be considered for.
Application procedure:Please include details for up to three referees, your full CV, your publication list and a 500-word research plan.
Application Procedure
You should submit your completed online application form at https://jobs.soton.ac.uk. The application deadline will be midnight on the closing date stated above. If you need any assistance, please call Sam Stubbs (HR Recruitment Team) on +44 (0) 23 8059 4043 or email recruitment@soton.ac.uk Please quote reference 1513821BJ on all correspondence.
More:
Lecturer in Data Science job with VICTORIA UNIVERSITY OF WELLINGTON | 275167 – Times Higher Education (THE)
M Te Herenga Waka - About our University
Te Herenga Waka - Victoria University of Wellington is a global-civic university with our marae at our heart. This iho draws off our heritage and is further defined by our trangawaewae, in particular Wellington, Aotearoa, and the Asia-Pacific, all of which are expressed in our position as Aotearoa New Zealand's globally ranked capital city university.
Our core ethical values are respect, responsibility, fairness, integrity, and empathy. These core ethical values are demonstrated in our commitment to sustainability, wellbeing, inclusivity, equity, diversity, collegiality, and openness. With, and as, tangata whenua, we value Te Tiriti o Waitangi, rangatiratanga, manaakitanga, kaitiakitanga, whai mtauranga, whanaungatanga, and akoranga.
Krero m te tranga - About the role
Te Herenga Waka - Victoria University of Wellington is currently recruiting a Lecturer in Data Science.
We seek a dynamic academic with a developing record of research in some branch or branches of a computational field within Statistics and Data Science. The successful applicant will contribute to the Data Science programme at Victoria University of Wellington, interacting with disciplines across the university who are users of Data Science, including bioinformatics, health sciences, and the humanities. The successful applicant will also contribute to the School's engagement with the wider Data Science community in the capital city and NZ-wide.
The appointee will undertake undergraduate and postgraduate teaching in data science at all levels. They will promote learning among students with a wide range of applied and computational domains within the subjects of Statistics and Data Science. They will undertake supervision of postgraduate research students on both theoretical and applied topics, including students in the workplace seeking to advance their qualifications. They will conduct an ongoing research activity of international standard, in keeping with the university's emphasis on high-quality scholarship and will contribute to building on the University's engagement with stakeholders, including through externally funded research.
pmanawa - About you
Our Ideal candidate will have the following:
The university is committed to building a diverse workforce and the School encourages applicants who are women, other gender minorities, Mori, Pasifika and from other groups underrepresented in the STEM sector.
tahi krero hai whina i a koe - Why you should join our team
The School of Mathematics and Statistics is well connected within the University, supporting opportunities to work with colleagues in biological, physical, health, social, earth and information sciences, engineering, business and the humanities. It also sustains strong connections with employers and researchers in the government, private and not-for-profit sectors, enabling work placements for students in our programmes.
Please go to our careers site to view the role description.
Contact details for vacancy: If you have any questions regarding this role please get in touch with A/Prof Ivy Liu, Head of School at ivy.liu@vuw.ac.nz but applicants should follow all steps listed below.
Important - Application steps and information
For applicants who are not NZ Citizens or Permanent Residents, we recommend you check the NZ Immigration website for updates related to Covid19 restrictions on entry to New Zealand:https://www.immigration.govt.nz/about-us/covid-19
Download and complete theUniversity Application Form.
Click the apply now button at the base of the advert, follow the process to enter your contact details and add your CV in the online form.
Please include a cover letter in your application, telling us why you're a great fit for this position then email the completed application form, cover letter and any other supporting documentation to erecruit@vuw.ac.nz stating the reference number and position title from the advert in the subject line.
Applications close Friday,25 February 2022.
Reference 3143.
Visit link:
Data science in the civil service: How to build capability to rewire government – Civil Service World
The Declaration on Government Reform set out a commitment to put data at the heart of our decision-making. There are already some pockets of data-excellence in government, thanks to many years work developing skills and capability both in centres of deep expertise such as the Office for National Statistics Data Science Campus, and through wider reforms such as rolling-out data training for perm secs.
But how do we move from the current landscape where the benefits of data and analytics are broadly understood but only play a key role in certain policy areas to the ultimate goal of a landscape where data is used not just to improve decision-making but to design better policies and processes right across government?
This session broughttogether data leaders from government and the private to ask how we can bridge the divide between governments data scientists and its policy and operational teams, building capability and skills across government to lead a data-driven revolution.
We considered what work has already been done as well as lessons government could learn from other sectors, with discussion on the latest thinking from academics and the private sector.
Fill in the form below to watch this webinar.
More:
The Top MIT SMR Articles of 2021 – MIT Sloan
The past years most popular articles offer insights on managing through an unpredictable environment of disruption and change.
In December 2020, the first highly anticipated doses of COVID-19 vaccines arrived, sparking hope that 2021 might bring a fresh new start less burdened by the pandemic. Since January, the world has made huge strides in managing the pandemic but coronavirus variants have advanced as well. Thanks to the delta variant, the Great Office Return expected in September seemed to peter out before it began, landing us at the end of another year characterized by constant adjustment, reorientation, and shifting plans.
In 2021, readers unsurprisingly gravitated toward articles about leading through a pandemic-changed world, understanding an unpredictable supply chain, and managing remote and hybrid teams. As the year progressed, readers focus shifted toward content about optimizing return-to-office plans, driving culture change, mitigating burnout, delivering on data science projects, and, of course, the always relevant issues of strategy and leadership. The following are 12 of the most widely read articles SMR published in 2021.
Get Updates on Transformative Leadership
Evidence-based resources that can help you lead your team more effectively, delivered to your inbox monthly.
Please enter a valid email address
Thank you for signing up
Privacy Policy
The shift to hybrid virtual and in-person work models requires a fundamental change in the skills team leaders need to succeed in 2022 and beyond. Leaders will need to play four roles Conductor, Catalyst, Coach, and Champion as they adapt to managing a hybrid post-pandemic workforce.
Todays leaders need best practices for dealing strategically and operationally with a distributed, diverse workforce that crosses internal and external boundaries. The best way to address the shift to managing all types of workers is through the lens of a workforce ecosystem a structure that consists of interdependent actors, from within the organization and beyond, working to pursue both individual and collective goals.
The Culture 500 research team studied the attributes that contribute to a strong organizational culture and found that 10 specific factors most influence employees positive or negative views of their companies. With the labor market upended by the pandemic, corporate leaders may need to pay attention to some previously invisible issues if they wish to retain valued employees.
The pandemic has had a major effect on the mental health and well-being of employees, with many reporting that they feel stressed and stretched too thin at work. Companies that dont help their people feel a sense of purpose, belonging, and progress amid these forces will see burnout persist or worsen. Leaders and managers can take seven specific steps to create a healthier work environment.
As organizations make their post-pandemic plans, they should take advantage of the opportunity to rethink how and where work is best done and how they can combine the best aspects of remote and colocated work.
Having the wrong culture undermines the best-laid strategy and organizational development plans, but many leaders have yet to be proactive in building the types of culture required for successful transformation. Describing seven elements of adaptive culture, the authors share a set of eight culture transformation principles that maximize the likelihood of success.
Organizational network analysis a methodology that maps employees working relationships can help guide decision-making for hybrid work plans and enable employees to improve their own effectiveness.
Its hard to understand why successful leaders suddenly fail to meet expectations what the authors term leader derailment. While personalities are sometimes to blame, organizational context plays a significant role. Companies can help prevent derailment by identifying the most challenging demands leaders must overcome early on and providing better support systems.
Employee data shows that people find virtual meetings draining because of their scheduling and structure. Leaders can make meetings more effective and less fatiguing by incorporating feedback from their teams.
Many companies are unable to consistently gain business value from their investments in big data, artificial intelligence, and machine learning. A study of the data science functions and initiatives in three of Indias largest private-sector banks identified five obstacles to successful data science projects and suggests remedies that can help companies obtain more benefit from their data science investments.
The ongoing global supply chain crisis shows no sign of abating. And although many media outlets have blamed pandemic shortages on companies practice of just-in-time inventory management, abandoning it would do little to help solve current supply chain problems.
Implementing strategy becomes a way of life when leaders follow six key practices drawn from examples of soulful business leaders.
Deborah Milstein is associate editor at MIT Sloan Management Review.
Visit link: