Tyler Malloy

Tyler Malloy

They/Them

Postdoctoral Researcher

Job Security

7 minute read

Date:

I recently completed the last meeting with a small group of mentees from my alma mater UBC that was put together to give undergraduate students at different points in their career some insight into life after undergrad. I think it was a great experience for myself, as I had little previous work as a direct mentor, and hopefully the mentees felt the same way. The remainder of this blog post is some reflections from my experience as a mentor and the life lessons I have thought of and tried to discuss with them.

The title of this post is partially a pun referencing the fact that I have recently begun work researching human and AI learning with a focus on applications in cybersecurity domains. If you had told me during my PhD that I would be working on that topic, I would have assumed that I left academia and began work in industry, finding the first relevant job that I could get. Interestingly, my impression of the differences between industry and academia in terms of job security and how much control one has over the type of work they do has evolved in the past year since defending my PhD (and I still haven’t even received my robe at commencement yet).

One of the often discussed (potential) benefits of staying in academia after completing a PhD is that one has more control over the type of work they do. I think that oftentimes this is assumed by PhD students, as well as people who never entered academia in the first place, to be a general statement about control at all levels, both the high motivation level and the lower applied level. However, the control that one has over the type of work you do is often different across these levels, and there is no one-size fits all description of how researchers in academia have control over their work compared to researchers in industry or government.

Since I have begun work in security, I am fascinated and amazed at the potential for using it as a domain for better understanding the similarities and differences between human and AI learning. For me, I was able to keep one aspect of the high level motivation of my research interests, while shifting the application to fit a new (and amazing) team of research collaborators. Again, for me, this was a somewhat easy change, since I had not been dead-set on a specific application of my research interests during my PhD. Part of my work initially was in robotics applications, and then visual learning and decision making. This meant that I was a good fit for the position that I applied for, and will (hopefully) be a good addition to the future research of this group.

The other meaning of the (punny) blog post is in reference to the recent downturn in hiring at large computer science companies. While I am far from an expert in determining the job market for different sectors, this was partially a reminder that nothing is guaranteed, and choosing one path over another because of a perceived benefit might not always be the best. For me, I could have easily left academia in pursuit of an industry job only to find myself out of work soon after. But since I chose to stay, there is no way to ever know what would have happened.

However, my story is just that, my own, and doesn’t mean that other people who do postdocs or researcher positions in academia have or want to have the same level of control over their research. Many researchers want to totally shift both the domain and application of their research, even significantly changing the area of science they study. Others only care about the low level aspects and want to change some way that research is applied to some specific real world application. These are all possible results of pursuing a PhD, and oftentimes the preferences that one has for the type of research control they have in the future is a fluidly changing thing.

This is all to say that I am very happy with my current position working in cyber security and other applications of human and AI learning. While I could have never guessed I would be working on this topic, at every step of my journey I have tried to do what seemed to me like the best options in relation to my preferences for both my life and research. I think that oftentimes people hold strongly to what they perceive as their preferences for their work or the future, and this can limit them in what they are willing to explore. This is especially true in research, which for the sake of good science shouldn’t be the case.

Also since the remaining aspect of the (punny) title is a reference to the doom and gloom over chat gpt replacing everyone’s jobs, I have again asked chat gpt to review my blog post and it had the following to say:

Chat GPT: Overall, this blog post provides interesting reflections on the author’s experience as a mentor and their personal journey transitioning from academia to industry, with a focus on their current work in cybersecurity and AI learning. The post also touches on the importance of being open-minded and adaptable when it comes to career choices and preferences. The author effectively uses a pun in the title to introduce the topic of their work in cybersecurity and AI learning, and explains how their research interests have evolved over time. They provide a nuanced perspective on the differences between control over one’s work in academia versus industry, highlighting that it is not a one-size-fits-all description. The author’s personal story adds a relatable and authentic touch to the post, and their reflection on the potential downsides of holding strongly to perceived preferences for work and the future is a valuable lesson. One potential area for improvement is that the post could benefit from more concrete examples or actionable advice for the mentees or other readers. The post is largely focused on the author’s personal experience and reflections, but it would be helpful to have some practical tips or recommendations for those looking to navigate their own career paths. Overall, this blog post provides an engaging and thought-provoking perspective on career choices and the importance of staying open-minded and adaptable.

Tyler: Interestingly, it seems to be under the impression that I have left academia for industry! I was trying to give just the opposite impression in my blog post. Also, the recommendations for improvements seem somewhat unrelated to what I was trying to describe with the post, as the general theme is that direct recommendations may not be useful due to changing preferences and life plans. Maybe jobs are a bit more secure than it would seem from the news.

You May Also Enjoy

Dynamic Decision Making Lab Yearly Review

2 minute read

Published:

The Dynamic Decision Making Laboratory has issued it’s yearly review for 2024 which summarizes the work from the lab that occured during the past year. It is available online here. I have copied below my own section of the yearly review, and encourage people interested in my work to check out the annual summary. This will also be of interest to those who would like to get an idea of the work that goes into running a relatively large research lab in decision sciences.

Descisions From Description and Experience

8 minute read

Published:

One of the interesting lines of recearch that I have recently become more involved with is the distinction between making decisions based on description and making them based on experience. In this research, these are referred to as Decisions from Experience (DfE) and Decisions from Description (DfD). In the past, I have studied differences and similarities between learning and decision making, which is as similar area of research in cognitive psychology. In the future I will be exploring whether these differences between decisions made from description and experience also occur in large language models, but for this post I will be focusing on the background of these differences.

Measures of Similarity

6 minute read

Published:

This post is inspired by the recent preprint that I released called ‘Leveraging a Cognitive Model to Measure Subjective Similarity of Human and GPT-4 Written Content’. In this work, we develop a method for estimating human’s subjective similarity using a cognitive model. Similarity between stimuli has been a major theme in my research, and it is a major challenge to cognitive modeling in general. In this post I will detail the importance of metrics of similarity that accurately reflect how humans perceive the similarity of stimuli they are presented with. I will focus on Instance Based Learning models, and in a later post discuss how similarity is relevant for Reinforcement Learning models and how we apply Generative AI representations onto determining similarity.

Multi-University Research Initiative

26 minute read

Published:

This blog post contains my unedited notes that I took while attending the Multi-University Research Initiative Program Review this year. This was an interesting meeting where I presented my work in cognitive models applied onto large languge model to improve their usefullness in educational settings, specifically for phishing email education.