Jekyll2026-03-10T17:17:26+05:30https://subhodip123.github.io/feed.xmlHomepersonal descriptionSubhodip Pandasubhodipp at iisc dot ac dot in✍️ Leisure Readings2024-08-15T00:00:00+05:302024-08-15T00:00:00+05:30https://subhodip123.github.io/posts/2024/08/blog-post-3This post accounts the non-academic books that I love reading during my free time. I mostly like reading non-fiction books with doses of fiction occassionally. Below are the books I have read over the years.

]]>
Subhodip Pandasubhodipp at iisc dot ac dot in
✍️ Navigating through Mistakes!2023-08-10T00:00:00+05:302023-08-10T00:00:00+05:30https://subhodip123.github.io/posts/2023/08/blog-post-2In the world of academia, accomplished researchers often appear as paragons of success, their profiles adorned with accolades and achievements. However, behind the facade of triumph, a less celebrated narrative existsβ€”a story of relentless struggle and failure that is seldom shared. As a fellow young researcher, I’ve keenly experienced the weight of this unspoken reality, often feeling isolated in my struggles. The purpose of this blog is to remind myself of my failures and also to make upcoming young researchers aware about the struggle. And Yes!! The struggle is real (atleast for me).

My voyage into research began in January of this year, and it has been far from smooth sailing. I can’t help but acknowledge the various challenges and moments of self-doubt that have become my constant companions. The initial problem statement I was excited to work on has taken an unexpected detour, and the experiments I thought would yield groundbreaking results have left me feeling somewhat underwhelmed. Upon introspection, I’ve identified a few key missteps that might have contributed to my current sense of failure. Here’s a candid exploration of my missteps that I want to share to other researchers also:

  • Lack of Extensive Literature Review: I’ll admit I was eager to dive into the practical aspects of my research, neglecting the wealth of knowledge that already exists in my chosen field. I came to realize that research isn’t just about creating something new; it’s about building upon what others have already discovered.

  • Not Clarifying the Significance of the Problem: It’s not enough to have an interesting problem fefining the significance of the problem statement is crucial. It’s about asking the β€œso what?” question.

  • Lack of Self-Evaluation: I have come to realise that healty quantity of self-evaluation is really important, especially when you’re venturing into uncharted territory. It’s a chance to refine my concepts, gather feedback, and iteratively improve my approach.

  • Not Sharing my Ideas: I also regret not actively reaching out to my colleagues and friends for their perspectives and insights. Sharing your research challenges with others can lead to fresh ideas and alternative viewpoints.

My current struggle and sense of uncertainty have led me to question my decision to leave a stable job for the tumultuous world of research. Is this what research is all about? A constant battle to identify meaningful problems and generate innovative solutions? There are days when I wake up without a clear sense of direction, and I wonder if I made the right choice. But then, there are moments of clarity and inspiration when I realize why I embarked on this journey in the first place. I think my β€œwhy” is purely a selfish self-exploratory reason to test by own abilites and enjoy.

As I navigate these uncharted waters, I’ve started to explore not only my β€œwhy” but also the β€œwhys” of others. There is a list of articles which I read now and then when I feel lost wandering:

]]>
Subhodip Pandasubhodipp at iisc dot ac dot in
✍️ Why of What intrigues me2023-07-10T00:00:00+05:302023-07-10T00:00:00+05:30https://subhodip123.github.io/posts/2023/07/blog-post-1Over the past century, the field of artificial intelligence (AI) has experienced remarkable progress, primarily driven by the aspiration to enable machines to learn in a manner akin to humans. This pursuit has been rooted in a fundamental question: β€œhow” can machines be effectively trained to perform specific tasks? It seems like the answer is the optimization of some specific loss functions, typically relying on carefully curated datasets.

However, as we embark on the journey toward achieving artificial general intelligence (AGI), a stage where machines are trained on extensive and diverse datasets, the central question shifts from β€œhow” to β€œwhat” these models are learning. It is no longer sufficient to focus solely on the mechanics of the learning process. Instead, we must delve into the profound inquiry of whether these AGIs possess intrinsic human values, such as privacy, safety, and fairness.

Furthermore, as we endeavor to endow AGI with human-like qualities, a critical attribute comes to the forefrontβ€”human awareness of the limitations of knowledge and the inherent uncertainty intertwined with it. Are these machines cognizant of their own uncertainty in decision-making, similar to human awareness, as our dependence on machine intelligence deepens? Can they promptly rectify their knowledge when exposed to erroneous information? Unfortunately, the answer to these inquiries is negative. These machines appear rigid and devoid of an understanding of the boundaries of their knowledge and the associated uncertainties.

To address these concerns, we must shift our focus from the β€œhow” to the β€œwhat” i.e. what these models learn from the data. As machine intelligence assumes an ever-expanding role across diverse fields, we are compelled to confront the fundamental question of β€œwhat” precisely they are assimilating and comprehending from the information to which they are exposed.

]]>
Subhodip Pandasubhodipp at iisc dot ac dot in