IIT Madras has invited applications for a free online course called Introduction to Machine Learning on the NPTEL platform. The course, which is AICTE FDP approved, can be taken by senior undergraduate or postgraduate students pursuing their BE, MS, ME or even PhD. The course is 12 weeks long and will be conducted from 26 July to 15 October 2021. It would be most beneficial for students pursuing education in the domains of computer science and engineering, artificial intelligence, data science, programming and robotics.

The course will be conducted by professor Balaraman Ravindran who is associated with the department of computer science at the Indian Institute of Technology Madras and is also a Mindtree Faculty Fellow. The professor has two decades of experience in machine learning and reinforcement learning. He is currently working in the areas of data mining, social network analysis, and reinforcement learning.

What the IIT Madras Free Online Course on Machine Learning Will Cover

Participants of the course will be introduced to some of the basic concepts of machine learning from a mathematically well-motivated perspective. Furthermore, the different learning paradigms and some of the more popular algorithms and architectures used in each of these paradigms will also be covered as part of the course. Some of the topics that students will get to learn are as follows:

• Probability theory, linear algebra, and convex optimization.
• Introduction: Statistical decision theory, regression, classification, and bias variance.
• Linear regression, multivariate regression, subset selection, shrinkage methods, principal component regression, and partial least squares.
• Linear classification, logistic regression, and linear discriminant analysis.
• Perceptron, and support vector machines.
• Neural Networks: Introduction, early models, perceptron learning, backpropagation, initialization, training and validation, parameter estimation, MLE, MAP, and Bayesian estimation.
• Decision trees, regression trees, stopping criterion and pruning loss functions, categorical attributes, multiway splits, missing values, decision trees, and instability evaluation measures.
• Bootstrapping and cross validation, class evaluation measures, ROC curve, MDL, ensemble methods – bagging, committee machines and stacking, boosting, and much more.

How to Enroll and Other Important Details

Interested students will have to enroll in the course before 15 October 2021. The course is free to enroll and learn from; however, those who require certificates will have to pay Rs 1000 and take an examination that will be conducted on 24 October 2021. Interested participants may enroll for the course on the official page by using their Facebook, Microsoft, Google or Facebook accounts.

The post IIT Madras Offers Free Online Course on Introduction to Machine Learning for Students appeared first on Artificial Intelligence.

NIT Warangal is currently accepting applications from students and professionals for a short-term online course on Python for Machine Learning. The course will be conducted from 26 to 30 July 2021 in the evenings around 5 pm for three hours a day. The course will cost Rs 500 for faculty, post-doctoral fellows, research scholars, postgraduate and undergraduate students, and will cost Rs 1000 for industry participants.

The course is being conducted by NIT Warangal’s Department of CSE and the Department of Mathematics in association with the Center for Continuing Education. Participants will get an e-certificate upon successful completion of the course. Suppose a participant misses any class, the course material and session recordings could be shared with participants with the consent of the speaker.

What the NIT Warangal Online Course on Python for Machine Learning Will Cover

The online course aims at introducing participants to python programming constructs, and at imparting knowledge and training on the fundamentals of python with different aspects and applications to machine learning. The following topics will be covered by NIT Warangal in the online course:

• Introduction to Python: Installation, and Python Editors.
• Variables, Objects, Operators, and Primitive Data Types.
• Compound Data types: List, Tuples, Sets, and Dictionaries.
• Conditional Statements Loops: for, while, do while.
• Functions, and Building your own functions.
• Numpy: Multi-dimensional Arrays.
• Matplotlib: 2D and 3D plotting in python Regular Expressions.
• Scipy: Scientific library for python Pandas: Providing high-performance, easy-to-use data structures.
• SymPy: Symbolic mathematics and computer algebra. scikit-image: Collection of algorithms for image processing.
• scikit-learn is a collection of algorithms and tools for machine learning.

How to Apply for NIT Warangal Online Course on Python for Machine Learning?

Interested participants will have to submit online applications through a dedicated Google form provided by the institute, and then pay the requisite fee before 20 July 2021. On receiving the Google form and fee remittance receipt, participants will be sent confirmation of their participation through email by 22 July 2021. Applicants are advised to go through the course brochure in detail prior to submitting applications.

The post NIT Warangal Offers Online Course on Python for Machine Learning to Students at Rs 500 appeared first on Artificial Intelligence.

AI-based assessment and grading in schools is a part of the new normal.

Once upon a time, there were real classrooms with teachers strictly evaluating you during your pen and paper examinations. This might be a story that our future kids will listen to if the ‘new normal’ is planning to stay here. The Covid-19 pandemic brought rapid digital transformation and AI-driven automation into all industries. According to IDC, the worldwide revenues for the artificial intelligence market are forecast to grow 16.4% in 2021 to USD 327.5 billion.

The growing significance of AI is also visible in the education sector. The pandemic-induced shift to online classes impacted many conventional methods of educational institutions. Remote learning has increased the accessibility and efficiency of the education systems. The shift towards virtual classrooms demands strong support from disruptive technologies. There have been many reports about the role of AI technologies in remote classes and virtual education systems and the ease it brings to different administrative tasks. Let us focus on one of the aspects where AI has already made its mark but has also raised some concerns, that is AI in exam evaluation.

AI in Evaluating Academic Performance

Recent reports revealed that Delhi’s new state school board, DBSE will be employing AI-based continuous assessment and game-based assessment in their schools. A Hindustan Times report states that AI in the assessment process will provide real-time learning feedback to teachers and students will be given situations or involved in activities to evaluate their skills and understanding through games. Will this mean that AI algorithms will decide the performance based on the skills and strength of the students? Absolutely, and this is not the first-ever approach. Many educational institutions and universities have already employed AI for assessment and automated grading systems. Schools in China had already started experimenting with automated AI grading systems a few years back.

AI-based assessments and marking techniques can minimize human biases and enhance the speed of evaluation. Instant feedback to both students and teachers is another advantage of AI in evaluations. Pearson, the multinational publishing and educational organization boasts having many AI-based assessment systems in ELT, which provides unbiased and accurate results.

AI paper grading softwares is gaining attention since they can quickly grade papers and assignments without any human intervention. Machine learning and data analytics are the pillars behind the technology of automated assessment. AI and Machine learning algorithms learn from existing data and try to replicate human evaluation patterns with precision.

AI-driven online marking and assessment, automated grading, AI-assisted proctoring are redefining the education system, while reducing bias and frauds. The Telangana State Board of Intermediate Education in India had announced the use of AI for accuracy in their exam results considering the discrepancies in evaluation that led to the suicides of many students. According to the Economic Times report, the board happened to find various errors in the assessment of OMR sheets due to flawed technology.

Are the Grades Reliable?

Despite the claims that AI-based assessment and grading is accurate, fair, and bias-free, there have been many instances where it went totally wrong. An article in the Harvard Business Review reveals how AI grading systems by the International Baccalaureate Organization produced varying results from the predicted ones and the students went on a protest. And, the AI system apparently, just predicted grades on the data fed to the algorithm rather than actually evaluating papers. Another article on The Verge says how a virtual learning platform’s AI-based assessment method focused on specific keywords to determine marks. Most of us would have heard how an automated AI algorithm caused havoc in the UK amid the pandemic by providing biased and lowered A-level results to the students.

We can find more such discrepancies if we dig more. Do these incidents indicate that AI-based exam evaluation systems are not an ideal approach after all? Maybe we are still lagging behind in addressing the flaws of AI and other disruptive technologies. While feeding loads of data to these systems, it must be taken care that the data do not represent any errors. The virtual and remote learning scenario is here to stay and AI will have many positive impacts on the education system. The future of AI is intertwined with us and hence, AI in the evaluation of exams will benefit many if incorporated in the right way.

The post STUDENTS CAN NOW ARGUE WITH AN AI SYSTEM FOR EXTRA MARKS appeared first on Artificial Intelligence.

HIGHLIGHTS

• By offering personalisation, customisation, and enabling the use of virtual assistants, Machine Learning has become increasingly relevant for students more than before.
• ML can’t stay behind in the race considering the various benefits it offers and the wide scope it has.
• Machine Learning helps to upscale the e-learning process by upgrading the content and tailoring it as per the present requirement.

The education industry is one of the most thriving sectors in India. While the pandemic brought with it various challenges, it also helped in digital adoption across all businesses. This trend was especially witnessed in the education sector. The teachers and students had to cope up with the changing times and had to move the learning sessions to the online medium.

To ensure that the educational firms and the ed-tech platforms extend the best of the knowledgeable sessions to their students, they need to be armed with state-of-the-art technology. The key technological advancement that has brought about a transformation in learning for the students is Machine Learning. Having this technology in place enables efficient and swift processing of big data thereby extending superior quality education to the students.

Importance of machine learning in educational firms and ed-tech platforms

An extension of Artificial Intelligence (AI), Machine Learning (ML) provides the ability to the IT systems to interpret data at their end and self-educate basis the experience gained. Its primary objective is to grab hold of the required data, analyze it and accordingly create a problem and solution algorithm.

The best bit about this technology is that it does not require the assistance of humans or external coding. Every aspect of the process is covered by the machines only. A prominent example of this is when IT professionals communicate directly with their clients and provide them necessary solutions.

While machine learning is widely used in IT and e-commerce firms, it has acquired an important place in the education sector as well. It has now become the need of the hour rather than being a luxury. The most common ways in which we integrate machine learning with education are via learning analytics and educational data mining.

Siri, the virtual assistant is another soon to become essential part of the e-learning process.

As reported by Technavio, the global market share of online courses based on machine learning technology is expected to grow at a CAGR higher than 16% by 2021. This growth will be driven primarily by the affordable large data volume storage and its efficient processing. It will further help in making the e-learning process interactive and result-driven.

1. Customisation and personalisation

Machine Learning algorithms analyse the mannerism by which the students explore the information they are provided with. In case, the students have grasped the topic well enough, they move ahead. But if not, then ML takes them back and makes them go through relevant points again to make them understand the concept better.

They also keep track of the learning process of the students at an individual level. ML gives an edge to e-learning over classroom teaching by ensuring that the students learn effectively with the varied formats of content in an interactive, engaging, and techie manner.

2. Upgradation of content

Machine Learning helps to upscale the e-learning process by upgrading the content and tailoring it as per the present requirement. They analyse the quality of the content of the various courses offered and compare it with them. Then, they accordingly make changes in their courses to provide state of the art education curriculum to the students. This helps them give an edge over the players in the market as they pro-actively alter their courses to extend the best of the educational experiences to their students.

3. Ensures higher ROI

When ML is applied, students get personalised services. Additionally, Machine Learning provides real-time benefits to the students thereby leading to their increased adoption in the education system. The process of predictive analysis further helps keep track of the progress of every student and also enables their quick grading.

4. Provides instant solutions

When incorporated, Machine Learning enables the platforms to provide instant solutions to the queries of the students. This benefit of ML makes it increasingly accepted by the educational and ed-tech platforms.

5. Helps save time by taking up admin-related jobs

ML helps save the time of students by automating the admin-related jobs. This makes the students devote their time wholeheartedly to their studies and assignments and they won’t waste their time on the admin-related jobs.

Summing Up

By offering personalisation, customisation, and enabling the use of virtual assistants, Machine Learning has become increasingly relevant for students more than before. While it cannot replace the teachers, it is used as a system to support and assist the traditional teaching methods. Since, digitisation is present and proliferating, with it, the technological advancements will amplify too.

ML can’t stay behind in the race considering the various benefits it offers and the wide scope it has. It has indeed revolutionised the education sector and will lead to further upscaling of the industry thereby being a major contributor to its growth.

The post Explained: Here’s why students should study machine learning appeared first on Artificial Intelligence.

RAPID CITY, S.D. (KOTA) – There are 9.2 quintillion possible outcomes of the March Madness brackets. That is 18 zeros after the nine!

No one has ever verifiably picked a perfect bracket, but some students at the South Dakota School of Mines are hoping to be the first, with the use of an artificial intelligence technique.

“Guessing that is essentially impossible, you’ve got to use some sort of assisted educated guess,” said Trevor Bormann, a senior metallurgy and computer sciences student at SDSMT.

They are using a technique called neural networking that is similar to the way the nerves in the retina of our eyes process transmit information to our brains. Their bracket model utilizes basketball statistics and deep analytics, called KenPom, plus, attributes of all the teams to pick a winner.

“What we’re not doing, is we’re not picking individual games, per say,” said Dr. Kyle Caudle, a math professor at SDSMT. “The output of the neural network will predict how far a team will make it in the bracket.”

Not only are they using this season’s Ken Pom data, they are using information from years past.

“But we also use historical data,” s aid Dr. Randy Hoover, a computer science and engineering professor at SDSMT. “So, we use all the past brackets to train our neural network model. And so, we can look at teams from the historical data and see how far they’ve made it in the brackets, based on all of the statistics that they use.”

Past massive upsets actually help the modeling– take the #16 seed UMBC upset over #1 Virginia in 2018. Even though the outcome was so unlikely, the data point was added to the model.

“It basically helps it not fall in a pitfall because if you have our 20 years of data, and 1 always beats 16, it’s always going to prioritize that 1 seed over the other,” said Jackson Cates, a junior computer science student at SDSMT. “So it helps it not make a guarantee to pick that.”

“What are you guys doing with this model? Are you just using it to beat everyone in office bracket pools for the rest of your lives?” asked Anderley Penwell.

The four laugh and Bormann said they are competing.

“We’re using it to compete in a competition, called the MinneMUDAC Competition, where there’s a whole bunch of teams from around the Midwest area that are creating their own artificial intelligence and machine learning algorithms and competing to see who can get the best bracket,” said Bormann.

There are 9.2 quintillion possible brackets.

“If you took every single human on Earth, and conscripted all of them to fill out brackets, 24/7, and everyone completed a bracket a minute, it would take 2,500 years or so to finish all of them,” said Bormann.

The post SDSMT students use artificial intelligence, trying to pick perfect bracket appeared first on Artificial Intelligence.

CBSE has launched an artificial intelligence community for students across the country in collaboration with Intel India to create an AI-ready generation

CBSE in partnership with Intel India has launched an artificial intelligence community for students in its efforts to help build a digital-first mindset and support an AI-ready generation. The platform aims at encouraging students to build real-life social impact artificial intelligence solutions as well as to collaborate and share their creations with others in the community.

The AI Student Community (AISC) is open for all students of the Central Board of Secondary Education (CBSE) and also for students of non-CBSE schools. The platform has been created in line with the National Education Policy 2020, which emphasizes on the importance of artificial intelligence and highlights on the efforts needed to prepare students of the country for an AI-driven economy.

Benefits of Joining the CBSE Artificial Intelligence Community for Students

Students who join the platform will be access to free videos on artificial intelligence and python, introduction to AI, AI domains, AI ethics, self-directed python notebooks, and so on. Apart from that, the platform also offers various other benefits to students such as:

• Students will be able to learn artificial intelligence as a skill and its application for creating social impact projects through webinars with Intel AI-certified coaches and experts.
• Participants will be given access to AI learning resources curated from across the world.
• Students can also attend webinars and face-to-face boot camps to enhance their artificial intelligence skills.
• Students can participate in online challenges to test their knowledge and up-level themselves, and also share their experiences as blogs.
• The community will serve as a platform for participants to connect with other students from across the country, and discuss and deliberate on relevant AI topics in a moderator curated forum.
• National and international competitions will be organised for participants.

How to Join the CBSE Artificial Intelligence Community for Students

Interested students, who belong to classes 6 to 12, can join the platform by visiting the official page and then carrying out the following steps:

• Click on ‘Sign Up’ and fill in the required details such as name, name of school, gender, etc. and submit.
• A verification email will then be sent on the registered email-ID, using which students will have to verify their email address.
• After that, the student’s profile will be created and their registration confirmed as a member of the AISC.

Students, once registered, can log in anytime using their email address and password to access all the features of the CBSE Artificial Intelligence Community.

The post CBSE Launches Artificial Intelligence Community for Students: All You Need to Know appeared first on Artificial Intelligence.

The programme will allow participants to build ML models for autonomous driving applications, the company said in a statement on Monday

In a bid to help women students learn machine learning (ML) via a Cloud-based virtual racing simulator, Amazon Web Services (AWS) has launched the DeepRacer Women’s League-India 2021.

The programme will allow participants to build ML models for autonomous driving applications, the company said in a statement on Monday.

The event will begin with a regional elimination round in four regions (north, east, west, south), followed by a national elimination community race.

The grand finale is on April 21 that will allow participants to upload their models and run on the virtual track, which will be streamed live on twitch.tv.

The winner of the India event will participate in the International DeepRacer League, providing an opportunity to acquire and test their ML skills on a global platform.

As an autonomous racing league, the initiative offers a unique opportunity to compete on tracks virtually while diving deep into advanced reinforcement learning and upskilling with AI and ML.

Participants will have the opportunity to leverage the AWS machine learning and AWS DeepRacer course modules and several enablement sessions planned during the event.

“The programme is open for all female students currently enrolled with higher education academic institutions in India, and is designed to promote upskilling in machine learning (ML), as well as inclusion and diversity in the technology sector,” said AWS, the Cloud arm of Amazon.

According to a global study, India has a higher-than-average share of women with AI skillsets (at 22 per cent) compared to other countries.

“This event is expected to foster community learning and help with the early adoption of artificial intelligence and machine learning technology for a more diverse workforce of the future,” the company said.

The post Amazon Web Services launches DeepRacer Women’s League to help female students learn Machine Learning appeared first on Artificial Intelligence.

Artificial Intelligence (AI) is revolutionizing the way we live and work , so today I invite you to learn more about this topic by approaching it from three different and increasingly complex segments.

Broadly speaking, we can refer to AI as the simulation of human intelligence by machines . In other words, a discipline that tries to create systems capable of learning and reasoning like people .

Importantly, Artificial Intelligence is the most debated technology of the 21st century . Today it is widely used to solve complex problems and facilitate human tasks.

Do you still have doubts about what artificial intelligence is? So join me in developing the following segments.

Segment 1: Explanation for elementary students

Imagine a machine that organizes your closet the way you like it , or that serves refreshments to your friends during a party . That is, a robot that performs the activities that you do daily. Well, this is precisely what Artificial Intelligence makes possible. Through algorithms and mathematical functions, AI provides machines with intelligence similar to that of a human to perform everyday tasks such as playing soccer with you or even giving you some dance classes

Through Artificial Intelligence, machines can learn, reason and solve problems . Three capabilities that make the robot artificially intelligent.

Segment 2: Explanation for Professionals

Artificial intelligence works by combining large amounts of data with fast, iterative processing, as well as intelligent algorithms, which allow software to automatically learn and identify patterns . The goal of AI is to create systems that can function intelligently and independently of human beings AND perform tasks in the professional field for example in the construction of buildings or in factories automated by intelligent robots.

Artificial intelligence has subfields, among which the following stand out:

Natural language processing : it is the ability of computers to analyze, understand and generate human language, including speech. Some products in this area are Amazon Alexa, Google voice

• A neural network : it is a type of machine-based learning that is made up of interconnected units (such as neurons), which process and transmit information.
• Cognitive computing : here a human-type interaction with machines is sought. Using artificial intelligence and cognitive computing, the ultimate goal is for a machine to simulate human processes through the ability to interpret images and speech. Some applications in this area are autonomous vehicles in which companies in Silicon Valley California work such as Cruise , Zoox and Nuro.

Segment 3: Explanation for Scientists

To develop this segment I am going to focus on the fact that Artificial Intelligence allows creating machines that can solve problems as people do . The interesting thing here is to compare them with humans and understand how they interact with them. To do this, the following points must be taken into account:

It is worth mentioning that there are currently4 types of Artificial Intelligence , according to Professor Arend Hintze :

Purely reactive: they do not have the ability to form memories , therefore they cannot use their experience in decision making. These are machines that make predictions based on parameters. An example of this type isDeep Blue , the IBM-made supercomputer that defeated chess grandmaster and world champion Garri Kasparov in 1997.

Limited memory: machines of this type do store memories and use that “past” to make decisions , that is, they use the knowledge of their previous experiences to elaborate a response. However, this information is kept for a certain time, so their vision of the past is very “limited”. To put it in some way, they store only a limited number of recent experiences that they use to perform their calculations and act on the environment, they do not use that experience to learn it forever, as people would. An example of this type is autonomous cars.

Theory of mind : one of the objectives of Artificial Intelligence is to ensure that machines can emulate the learning process that people have. The machines of this classification are then more advanced, they are capable of processing and expressing emotions . Theoretical knowledge of these emotions allows them to know how people or elements in their environment think and feel, and to predict what we expect them to do, to later adjust their behavior according to those predictions.

Self-awareness : this would be the highest step in the hierarchy proposed by Professor Hintze . In it would be the machines that are aware of themselves, recognize their internal states and become aware of the place and position they occupy within an environment. In this case they could not only predict behaviors but also emotions because they would understand what these entail in all its complexity.

Amazing isn’t it?

As we can see in the three explanatory segments above, Artificial Intelligence is designed to work with humans and facilitate their daily tasks . We currently find it applied in: banks, online customer service, cyber security, virtual assistants, smartphones, cars, social networks, video games and in many other aspects of daily life.

This panorama allows us to see that in the near future we will no longer think about how to use emerging technologies, but we will have to learn to relate to them and live with machines in a natural way .

If you want to learn more about Artificial Intelligence, I invite you to be part of Ai Lab School , the first practical education program in AI programming in Mexico . The course lasts 9 months , where you learn, based on projects, to develop solutions using advanced algorithms with AI. Once the course is finished, you access the next stage: job placement . Here the Ai Lab School team guides and helps you to apply for job offers in Silicon Valley and in technology hubs in the United States, in collaboration with the Mexican talent connection program in innovation and technology companies and startups Talentum Space .

• Speech recognition , an important point today
• Object detection
• Solve problems and learn from the data provided
• Planning approach for future tests to be performed

The post What is Artificial Intelligence? Explanation to: elementary school students, professionals and scientists appeared first on Artificial Intelligence.

Online education when compared against the traditional method adds to the features, and aids teaching methods when AI is used properly.

Online education or not, students love interactive classes where they don’t feel saturated. Artificial intelligence can help create such an interesting, useful platform. The changed world does not limit the age group for learning, and hence there are innumerable and diverse ways of educating oneself through AI. Contrarily, implementation of technology in education also changed the format of teaching-learning experience. So what exact role does AI play in education, and despite the possible discrepancies, what should one expect from online education?

The first and foremost change everyone experiences is the shift of teaching method from linear curriculum to project/skill oriented learning. This ensures that the students would learn the soft skills such as problem solving, effective communication, etc essential in the work life in future. Technology is also said to help in increasing student engagement, although it does not guarantee so.

Better visuals and experiential learning through interactive learning applications help a lot in attracting the attention of students. The cloud feature of AI in virtual learning has the provision of revisiting the lectures and the study material, even after the day is called. Teachers can also personalize the learning experience for each students, as technology facilitates the progress tracking of each student.

Virtual classrooms have the perk of collective intelligence; students can share whatever they find relevant and interesting to a particular concept taught in the class. Again, the involvement in the class depends on the student in the end, all AI can assure is betterment of the chances of that happening. AI opens up many creative gates for both students and teachers. The assignments for students can be out of the box; showcasing their abilities and knowledge beyond the prescribed books, which in turn makes them more confident about their work. Teachers, then, could figure out the inclination of each student from a considerably young age.

This changes the whole dynamic of the education, where teachers can become mentors and guide the students through their careers. The virtual world also reduces expenses such as field trips, guest lectures, etc. Students get to experience various worlds, developing them in all directions and creating an awareness within them. Augmented reality is a branch undergoing advancement at an exponential rate, and it is expected to be included soon in online education.

Online education can also increase family engagement, an important factor in the development of students. Teachers are able to find better material for students online, and improve their performance of instruction through direct/indirect constructive feedback from both parents and students. There are many obstructions in this type of education as of now. There is unequal access to technological gadgets and teachers are yet to understand the best use of AI in teaching. Moreover, too much engagement with devices is leading to a lot of health issues.

Apart from this, implementing AI in our daily lives also require deep knowledge of it. Which means that the curriculum needs to be updated, including technological skills at school level. Students, or everyone, should learn the social sharing skills, sustainable digital presence, ways of data protection, and online collaborative skills for a better and safe future in this specific itinerary.

The post Role of Artificial Intelligence in Online Education appeared first on Artificial Intelligence.

When you take a medication, you want to know precisely what it does. Pharmaceutical companies go through extensive testing to ensure that you do.

With a new deep learning-based technique created at Rice University’s Brown School of Engineering, they may soon get a better handle on how drugs in development will perform in the human body.

The Rice lab of computer scientist Lydia Kavraki has introduced Metabolite Translator, a computational tool that predicts metabolites, the products of interactions between small molecules like drugs and enzymes.

The Rice researchers take advantage of deep-learning methods and the availability of massive reaction datasets to give developers a broad picture of what a drug will do. The method is unconstrained by rules that companies use to determine metabolic reactions, opening a path to novel discoveries.

The research by Kavraki, lead author and graduate student Eleni Litsa and Rice alumna Payel Das of IBM’s Thomas J. Watson Research Center, is detailed in the Royal Society of Chemistry journal Chemical Science.

The researchers trained Metabolite Translator to predict metabolites through any enzyme, but measured its success against the existing rules-based methods that are focused on the enzymes in the liver. These enzymes are responsible for detoxifying and eliminating xenobiotics, like drugs, pesticides and pollutants. However, metabolites can be formed through other enzymes as well.

“Our bodies are networks of chemical reactions,” Litsa said. “They have enzymes that act upon chemicals and may break or form bonds that change their structures into something that could be toxic, or cause other complications. Existing methodologies focus on the liver because most xenobiotic compounds are metabolized there. With our work, we’re trying to capture human metabolism in general.

“The safety of a drug does not depend only on the drug itself but also on the metabolites that can be formed when the drug is processed in the body,” Litsa said.

The rise of machine learning architectures that operate on structured data, such as chemical molecules, make the work possible, she said. Transformer was introduced in 2017 as a sequence translation method that has found wide use in language translation.

Metabolite Translator is based on SMILES (for “simplified molecular-input line-entry system”), a notation method that uses plain text rather than diagrams to represent chemical molecules.

“What we’re doing is exactly the same as translating a language, like English to German,” Litsa said.

Due to the lack of experimental data, the lab used transfer learning to develop Metabolite Translator. They first pre-trained a Transformer model on 900,000 known chemical reactions and then fine-tuned it with data on human metabolic transformations.

The researchers compared Metabolite Translator results with those from several other predictive techniques by analyzing known SMILES sequences of 65 drugs and 179 metabolizing enzymes. Though Metabolite Translator was trained on a general dataset not specific to drugs, it performed as well as commonly used rule-based methods that have been specifically developed for drugs. But it also identified enzymes that are not commonly involved in drug metabolism and were not found by existing methods.

“We have a system that can predict equally well with rule-based systems, and we didn’t put any rules in our system that require manual work and expert knowledge,” Kavraki said. “Using a machine learning-based method, we are training a system to understand human metabolism without the need for explicitly encoding this knowledge in the form of rules. This work would not have been possible two years ago.”

Kavraki is the Noah Harding Professor of Computer Science, a professor of bioengineering, mechanical engineering and electrical and computer engineering and director of Rice’s Ken Kennedy Institute. Rice University and the Cancer Prevention and Research Institute of Texas supported the research.

The post New deep learning-based technique could boost drug development appeared first on Artificial Intelligence.