From the Editor

N Ramamurthy

While many of us have heard about ‘natural’ and ‘artificial’ intelligence, what is making rounds today is the concept of ‘General Intelligence’. GPT is a Neural Network which is inspired by biological process in the brain that mimics the nature of interactions between neurons in the brain to achieve a wide range of tasks. For a while, interest in neural networks had waned and it seems to have got a fresh impetus due to availability of large amounts of data and the availability of powerful computers making it possible to train neural networks on large datasets.

In his article titled ‘GPT-3: An Important Step Towards General Intelligence’, the author dwells in this domain to bring out three concepts on which  GPT-3 Architecture is developed,  a number of potential applications of GPT in such applications as text summarization, development of alternative search engines to rival the popular Google search engine, Intelligent Chatbots  etc., and its potential misuse  by bad actors resulting from the high quality of text generated by GPT-3 which facilitates spam, phishing, misinformation campaigns, social engineering attacks etc., which were previously constrained by their poor quality of text. Given that GPT-3 is trained on a large corpus of text scraped from the Internet, where publishing information is largely unregulated (eg: Reddit channels) there exists concerns over the spread of misinformation about sensitive issues like for example COVID-19 pandemic or electoral processes, if it falls into the hands of threat actors.

Turning our attention to the next article, it is often heard that a large percentage of graduates coming out of college/universities are unemployable simply for the reason that whatever the knowledge they have acquired over several years of education does meet the current needs of the industry. Most employers, recognise that graduates with a demonstrably wide range of skills are more likely to be better prepared to tackle complex problems while also understanding the needs, desires, and motivations of co-workers, clients, and the wider society. A holistic and multidisciplinary education seeks precisely these outcomes – strong written and oral communication skills, teamwork skills, ethical decision making, critical thinking, and the ability to apply knowledge in real world settings. The National Education Policy 2020 (NEP2020) is expected to herald a new era in this regard. Knowing that every graduate entering the job market today will look forward not only to several jobs, but also several careers during their working life, the need for a wider range of skills is evident. A holistic and multidisciplinary education has the potential to provide graduates with a combination of transferable and uniquely human skills, to help them adapt and continuously learn to work in this challenging environment.

In the article titled ‘Holistic and multi-disciplinary engineering education: why and how?’, the author focuses his  attention more specifically to Engineering education. Designing holistic and multidisciplinary Engineering education program is obviously not a simple matter of offering a set of unrelated courses, or even courses where some lectures examine a central theme from the perspective of one discipline and other lectures examine the same theme from the perspectives of other disciplines. His article has provided answers to the questions of ‘why and how’. The author concludes that it is now up to the leading engineering institutions to take a lead in implementations of these ideas, and to create models for other Engineering institutions to adopt.

The above two articles apart, in a continuing series on the tutorial we have on Gamified Approach to Learning Algorithms (GALA), the author takes the example of Huffman Codes where the challenge lies in variable length encoding that uses a shorter code for more frequently used cases, and longer codes for less frequently (length of code in some way is inversely related to frequency of occurrence), and with the additional requirement that no code should appear as a prefix of another code.

Hope the above set of articles to be interest to a wide range of our readers.