Wordle for AI: Santiago Valderrama on Getting Smarter on Machine Learning

Want to learn about AI and machine learning? There are plenty of resources out there to help — blogs, podcasts, YouTube tutorials — perhaps too many.

Machine learning engineer Santiago Valdarrama has taken a far more focused approach to helping us all get smarter about the field.

He’s created a following by posing one machine learning question every day on his website bnomial.com.

Think of it as Wordle for those of who want to learn more about machine learning.

As Valdarrama wrote on a LinkedIn post: “I got together with a couple of friends and built bnomial — a site with a simple goal, a non-BS simple way to learn something new as fast as possible. We published one machine learning question every day. That’s it. You load the page, answer the question and return the next day. Rinse and repeat.”

NVIDIA AI podcast host Noah Kravitz spoke with Valdarrama to talk to him about binomial, how to get smart about machine learning, and his own journey in the field.

Read more : 

www.dprg.com

How machine learning model from IIT-Madras team could boost personalised cancer therapy

IIT-Madras study, published in the journal ‘Frontier in Genetics’, dwells on algorithm to identify personalised genes that have the potential to form and drive cancer in individuals.


Researchers at Indian Institute of Technology (IIT)-Madras have developed a machine learning (ML) algorithm to identify personalised genes that have the potential to form and drive cancer in individuals. The model uses a ‘multiomic’ approach, the combined study of intersectional studies that end with the suffix ‘-omics’.

Details of the algorithm were published in a peer-reviewed paper in the journal Frontier in Genetics last month. The findings are expected to help in devising more personalised cancer therapies, contributing to the growing field of targeted therapy and immunotherapy trials.

Called ‘Personalized Identification of driVer OGs and TSGs’, or PIVOT, the model identifies personalised drivers of cancer genes and classifies them as either tumour suppressor genes (TSG) or oncogenes (OG) — the two types of genes involved in cancer.

The algorithm has also identified rarer driver genes that have not been studied enough to be associated with some cancers in large pan-cancer databases.ML works by consuming large datasets and understanding previously identified patterns. It then applies logic to identify new patterns in new (or existing) data.

For this study, the IIT-Madras researchers worked on datasets that contained genomes of individuals with four kinds of cancers — breast cancer, colorectal adenocarcinoma, lower grade glioma (brain tumours), and lung adenocarcinoma — and their identified driver genes and mutations.


The model classified genes as neutral or drivers, and further labelled them into TSGs and OGs.

It also identified other trends in data and newer driver genes from medical literature.

“It is difficult to predict how well we can train and predict genes in unseen cancer types,” said Malvika Sudhakar, lead author of the paper, to ThePrint. “A lot of factors, such as the number of samples, influence the performance of the model.”

Read more :

www.dprg.com

LaMDA: The hype about Google AI being sentient

The article has been authored by Sanur Sharma, an associate fellow, Manohar Parrikar Institute for Defence Studies and Analyses.


Artificial Intelligence (AI) has been considered the key to the future when it comes to imitating the human brain or becoming sentient. Recently, Google’s AI engineer Blake Lemoine going public on Google LaMDA, has sparked a discussion on AI models achieving consciousness. But what is more important here with these sparks is the serious concern regarding AI ethics.

So, what exactly is LaMDA, and why it is called sentient?

LaMDA is Google’s Language Model for Dialogue Applications. It is a chatbot based on the big advanced language model that can ingest trillions of words from the internet to inform its conversation. It is built on a massive corpus of data or text crawled from the internet. It is a statistical abstraction of all the text. So, when this system or model is asked, it takes the text written in the beginning, tries to continue based on the words related to one another, and predicts what words it thinks will come next. So, it is a suggestive model that continues to the text you put in. LaMDA has similar skills to BERT and GPT-3 language models and is built on Transformer, a Neural Network architecture that google research invented in 2017. The model produced through this architecture has been trained to read words, sentences and paragraphs, relate words with one another, and predict words that would come next in the conversation.

So how is it different from other chatbots also designed for conversations? Chatbots are conversational agents meant for specific applications and follow a narrow predefined path. In contrast, according to Google, “LaMDA is a model for dialogue application capable of engaging in free flow conversations about seemingly endless topics”.

The general characterisation of conversations tends to revolve around specific topics, and due to their open-ended nature, the conversation can end up in a completely different domain. According to Google, LaMDA is trained to pick up these several nuances of language that differentiate open-ended conversations from other forms making them more sensible. The Google 2020 research states that “Transformer based Language Model based on Dialogue could learn to talk about virtually anything”. It further stated that LaMDA could be fine-tuned to improve its sensibleness and specificity of the response

Read more : 

www.dprg.com

IIT Madras Develops Artificial Intelligence-based Tool to Predict Cancer-causing Genes

The Indian Institute of Technology (IIT) Madras researchers claims to have developed an artificial intelligence-based tool, called ‘PIVOT’, that can predict cancer-causing genes in an individual. This tool aims to help in devising personalised cancer treatment strategies, the institute said. The researchers added that the tool is based on a machine learning model that classifies genes as tumour suppressor genes, oncogenes or neutral genes.



Explaining the use of newly developed tool to treat cancer, researchers claimed that it is an uncontrolled growth of cells that can occur due to mutations in oncogenes or by tumor suppressor genes or both but not all mutations necessarily result in cancer. Therefore, it is important to identify genes that are causing cancer to devise appropriate personalised cancer treatment strategies.

“PIVOT is designed to predict genes that are responsible for causing cancer in an individual. The prediction is based on a model that utilizes information on mutations, expression of genes, and copy number variation in genes and perturbations in the biological network due to an altered gene expression," IIT Madras said.

The IIT Madras researchers said that they have built AI prediction models for three different types of cancer including breast invasive carcinoma, colon adenocarcinoma and lung adenocarcinoma. The team is also working on a list of personalised cancer-causing genes that can help in identifying the suitable drug for patients based on their personalized cancer profile.

The research was led by Prof Raghunathan Rengaswamy, Dean (Global Engagement), IIT Madras, and Professor, Department of Chemical Engineering, IIT Madras, Dr Karthik Raman, associate professor, Bhupat and Jyoti Mehta School of Biosciences, IIT Madras and a core member, Robert Bosch Centre for Data Science and Artificial Intelligence (RBCDSAI), IIT Madras, and Malvika Sudhakar, a research scholar, IIT Madras.

Highlighting the significance of the research, Dr Karthik Raman said, “Cancer, being a complex disease, cannot be dealt with in a one-treatment-fits-all fashion. As cancer treatment increasingly shifts towards personalised medicine, such models that build toward pinpointing differences between patients can be very useful.” “The research area of precision medicine is still at a nascent stage. PIVOT helps push these boundaries and presents prospects for experimental research based on the genes identified,” Malvika Sudhakar, Research Scholar, IIT Madras said.

More info: 

www.dprg.com

Canadian researchers using machine learning to mitigate effects of climate change

After spending almost a decade working in computer science and artificial intelligence (AI), Sasha Luccioni was ready to uproot her whole life three years ago after she became deeply concerned by the climate crisis.

But her partner convinced her to not give up her career completely but instead apply her knowledge of AI to some of the challenges posed by climate change.

"You don't need to quit your job in AI in order to contribute to fighting the climate crisis," she said. "There are ways that almost any AI technique can be applied to different parts of climate change."

She joined the Montreal-based AI research centre Mila and became a founding member of Climate Change AI, an organization of volunteer academics who advocate using AI to solve problems related to climate change.

Luccioni is part of a growing community of researchers in Canada who are using AI in this way.

In 2019, she co-authored a report arguing that machine learning can be a useful tool for mitigating and adapting to the effects of climate change.

Computer scientists define machine learning as a form of artificial intelligence that enables computers to use historical data and statistical methods to make predictions and decisions without having to be programmed to do so.

Common applications of machine learning include predictive text, spam filters, language translation apps, streaming content recommendations, malware and fraud detection and social media algorithms.

Applications for machine learning in climate research include climate forecasting and optimization of electricity, transportation and energy systems, according to the 2019 report.
Preparing for crop diseases

Researchers at the University of Prince Edward Island (UPEI) are using AI modelling to warn farmers about risks to their crops as weather becomes more unpredictable.
Drones, AI being used in climate change research at new centre on P.E.I.

"If you have a dry year, you see very little disease, but with a wet year, you can get quite a bit of disease around plants," said Aitazaz Farooque, interim associate dean of UPEI's School of Climate Change and Adaptation.

Read more : 

Some parts of the world are investing more heavily in data analytics roles than others


Asia-Pacific was the fastest growing region for data analytics hiring among packaging industry companies in the three months ending May.

The number of roles in Asia-Pacific made up 3% of total data analytics jobs – up from 1.5% in the same quarter last year.

That was followed by South & Central America, which saw a 1.4 year-on-year percentage point change in data analytics roles.


The figures are compiled by GlobalData, who track the number of new job postings from key companies in various sectors over time. Using textual analysis, these job advertisements are then classified thematically.


GlobalData's thematic approach to sector activity seeks to group key company information by topic to see which companies are best placed to weather the disruptions coming to their industries.

These key themes, which include data analytics, are chosen to cover "any issue that keeps a CEO awake at night".

By tracking them across job advertisements it allows us to see which companies are leading the way on specific issues and which are dragging their heels - and importantly where the market is expanding and contracting.

Which countries are seeing the most growth for data analytics job ads in the packaging industry?

The fastest growing country was Brazil, which saw 0.2% of all data analytics job adverts in the three months ending May 2021, increasing to 1.6% in the three months ending May this year.

The top country for data analytics roles in the packaging industry is the United States which saw 80.4% of all roles advertised in the three months ending May.
Which cities and locations are the biggest hubs for data analytics workers in the packaging industry?

Some 9.3% of all packaging industry data analytics roles were advertised in Fairborn (United States) in the three months ending May.

That was followed by Westminster (United States) with 8.1%, Broomfield (United States) with 5.4%, and Columbia (United States) with 3.9%.

More info : 

www.dprg.com

FICO Named Best Technology Provider for Data Analytics at 2022 Credit Awards

FICO has developed a unified FICO® Platform for decision management that includes capabilities for data management, analytic development and execution, strategy design and reporting. These capabilities can be shared across different functions, giving users a 360-degree customer management tool. Leading banks across Europe are adopting FICO Platform in order to break through the silos in their organization and propel growth while reducing the costs associated with installing and maintaining different point solutions for each area of the business.

One important capability set that is growing in usage across EMEA is prescriptive analytics, or optimization. This involves the use of advanced, AI-powered analytics to determine customer strategies that meet specified business goals under constraints. Banks across Europe have seen dramatic increases in performance by using FICO optimization to grow the business, manage risk and improve customer satisfaction.

The strength of FICO Platform and its embedded optimization capabilities was acknowledged in late 2020 with The Forrester Wave™: Digital Decisioning Platforms, Q4 2020, an industry analyst report that named FICO as category leader.

“The judges felt FICO was industry-leading in terms of driving change through a single decisioning platform,” said Luke Broadhurst, CEO of Shard Media, which publishes Credit Strategy magazine. “It offered an impressive set of metrics highlighting impact on the customer, with evidence of improved performance and customer outcomes.”

“We are celebrating this achievement, which validates the strength of our vision for a unified decision management platform,” said Matt Cox, vice president and general manager of FICO in EMEA. “From data ingestion to predictive modelling to prescriptive analytics and decision optimization, our platform gives our hundreds of clients across EMEA a real analytic advantage.”

About the Credit Awards

The Credit Awards are known as the leading awards in EMEA for credit professionals, with 30 categories this year. The awards recognise and celebrate innovation, best practice and the hard work of individuals, business divisions and pan-global conglomerates across the entire industry. The 2022 awards were judged by 16 leading figures in the EMEA credit industry.

About FICO

FICO (NYSE: FICO) powers decisions that help people and businesses around the world prosper. Founded in 1956, the company is a pioneer in the use of predictive analytics and data science to improve operational decisions. FICO holds more than 200 US and foreign patents on technologies that increase profitability, customer satisfaction and growth for businesses in financial services, telecommunications, health care, retail and many other industries. Using FICO solutions, businesses in more than 120 countries do everything from protecting 2.6 billion payment cards from fraud, to helping people get credit, to ensuring that millions of airplanes and rental cars are in the right place at the right time.

Learn more at www.fico.com

Data Science Bootcamps: What You Need To Know

As companies have embraced data analytics over the past 20 years, demand for data science professionals has grown drastically. Many people assume that a traditional degree is the only way to get a data scientist job, but a data science bootcamp can be an excellent alternative.

As this article will discuss, data science bootcamps are one of the best ways to learn applicable skills in a short amount of time.

What Is a Data Science Bootcamp?

If you’re considering a data science bootcamp, you should first understand what to expect from one of these programs. Some professionals think bootcamps are similar to college, but that isn’t necessarily the case.

For instance, most data science bootcamps are only three to six months long and focus on project-based learning. Instead of multiple classes on theory and information, bootcamps focus on teaching students in-demand industry knowledge and skills that they can apply to their jobs on day one.

Data science bootcamps also emphasize flexibility. A variety of programs accommodate different learning styles and schedules. For example, bootcamp students might study full or part time. They might prefer either in-person or virtual learning as well. Depending on which option you choose, program length and costs may vary.

Bootcamp graduates can choose among many paths when looking for their first job in the industry.

Common roles for data science bootcamp graduates include:Data scientist

Data analyst
Business analyst
Data engineer
Database administrator

Who Should Attend a Data Science Bootcamp?

Bootcamp students come from various backgrounds. Some currently work in the technology industry, and others are looking for their first job in the sector.

For this reason, many data science bootcamp programs teach the fundamentals first before diving into more complex skills. This equips all learners with the same foundational skills needed for success in a bootcamp.

How Much Does a Data Science Bootcamp Cost?

A traditional undergraduate degree in computer science could cost anywhere between $15,000 and $30,000 per year. In contrast, the average cost for a full bootcamp is $14,000.

For potential students who are hesitant about the cost of attending a data science bootcamp, most programs offer financing options to help offset the price. For example, BrainStation allows students to pay for the program in installments instead of a lump sum. BrainStation also offers several scholarship opportunities for qualifying students.

Talk to your prospective bootcamp’s admissions team to see whether you qualify for financial aid or financing offerings.

How to Enroll in a Data Science Bootcamp

Make sure to review several data science bootcamps before settling on one. Each bootcamp offers its own experiences, projects and lessons, so it’s important to choose a program that suits your needs.

When considering bootcamp options, consider the following factors:Experience of instructors
Curriculum
Portfolio work opportunities
Post-graduation support
Cohort makeup

Each of the above factors affects bootcamp outcomes. Ultimately, instructors are the most crucial part of any bootcamp. When comparing instructors, look for teachers who have worked in the industry and know what skills are necessary to succeed as a data science professional.
Are there Prerequisites that Need to be Met to Enroll in a Data Science Bootcamp?

Most data science bootcamps do not involve prerequisites. However, given how fast-paced and intensive these programs can be, it may be helpful to build a basic understanding of data science before beginning a bootcamp.

To prepare yourself for the first day of class, consider brushing up on basic statistics and intermediate mathematical computations.

What a Data Science Bootcamp Teaches You

Compared to a traditional college education, data science bootcamps don’t teach much theoretical knowledge. Instead, bootcamps focus on developing applicable skills and familiarity with technologies that graduates will use daily.

The topics you learn and how you learn them depends on which data science bootcamp you choose. Most data science bootcamp students will explore some variation of the ideas listed below.
Coding Languages

Individual data science bootcamps may teach multiple languages such as Java or C++, but most programs focus on Python.

Python is a versatile programming language used for many tasks, including website development and machine learning. Python is also an excellent tool for data scientists to quickly organize and analyze large data sets.

Depending on your instructor, you may also learn additional Python programming tools such as Python libraries.
Machine Learning

Machine learning is another focus area for many bootcamps. With machine learning, you’ll be able to set up computers to automatically perform tasks without programming. Data science students typically learn skills like regression analysis and logistic regressions to help perform machine learning tasks.
Data Science Fundamentals

Most individuals who start a data science bootcamp are excited to jump right into coding, machine learning or “big data.” However, understanding the basic fundamentals is essential if you want to build a lasting career.

During the first week of a bootcamp, instructors often teach students how to utilize probability theory and run A/B tests. These basic skills make it easier to tackle larger fundamental tasks later in the program.
Soft Skills

Bootcamps focus on developing technical skills, but they may also teach certain soft skills. For instance, data scientists aim to solve problems using information. Data science bootcamp students complete projects that help build their problem-solving skills and allow them to develop their own strategies for addressing issues.

You might learn the following soft skills in a bootcamp as well:Written communication
Verbal communication
Networking skills
Team

More info : 

www.dprg.com

The average data scientist earns almost $100,000 a year — and the barriers to entry for candidates are being broken down

Companies are facing a talent shortage of experienced data scientists due to evolving technology and inflated salaries.

The need for data specialists has moved beyond the traditional tech roles. The number of jobs requiring data science skills is projected to grow by 27.9% by 2026, according to the US Bureau of Labor Statistics.

Matthew Forshaw, senior advisor for skills at The Alan Turning Institute, said his research into data skills in the UK found there was a growing demand for professionals in the finance, insurance, and manufacturing sectors.

The increased demand is putting a recruiting squeeze on the entire sector, with companies reporting they are struggling to find experienced candidates.


It's a new discipline, it's been changing quite a lot," Libby Kinsey, head of data science at Ocado, a UK-based company that licenses grocery technology, told Insider. "So it's just quite hard to find the right people with the right skills."

"I would say the shortage is a lot on the leadership side and very senior people," Claire Lebarz, head of guest data science at Airbnb, said. "And the salary war we're seeing doesn't help at all."

Data scientists in the US make an average base salary of $97,000 a year, according to Payscale. But senior data scientists at top companies can make more than double this average. Data scientists at Facebook's parent company, Meta, for example, can make up to $260,000 in base salary a year, according to disclosed foreign labor data.

Experienced data scientists and industry experts told Insider what they are looking for in new recruits.

More info :