Видео

Please give us the literature review tutorial

Andrey, thank you for the video!

Журнал то ниже среднего. Картинки типичные советские. Можно было сразу сказать что никто не будет цитировать такой некондиционный продукт. Какой же это топ журнал?

A template for how to make sure your paper gets cited a lot, if your whole research group follows it. Problem A has been gaining more and more attention recently (citations by your groups and their friends). Many possible solutions have been proposed (citations by friendly groups). One of the more promising solutions is B (citation where idea is introduced/explained). B has been promising in several context (citations to applications of B by your own group). B has these advantatges over those other solutions (other citations by your own group). Recent developments of B has been focussed on C, D, E (citations of your own work or people you collaborate closely with). However, B still lacks minor thing F (citation to your last paper). Therefore, in this paper, we will ....

tenet moment

All your points are valid. Why do you think conferences are for deep scientific discussions? I truly value the deep discussions, but I’ve never saw them on big conferences nowadays. I see that modern conferences are for presenting your work, yourself and spending time with colleges that you already know. If somebody presented they’re done. They are out of attention. It’s like you have passed the exam and doing your things. I don’t like it also. However сonferences started to have an exam vibe. The best talks I had was after a conference in a bar with an alcohol. 😅 Why do you have a conviction that conferences are for discussions?

A complete ATAS submission means that upon submission you will get 6-digit Reference Number and Green color "Submitted" written infron of 6 digit ATAS reference number. And in case if anyone is getting Orange color ATAS ref number which is of 10 digit,it means that you have not yet submitted the application to ATAS Team. and to make it correct you have to change the course start date.which means if it is 12 dec today then you have to mention 13 dec course start date(the next day date) on option 1 that is Propose Study and then "CLICK ON" left side menu which says "Go To Part 2 of 10". Thats how you can get te 6-digit ATAS number.

Adam: A Method for Stochastic Optimization It's one of my favourite paper titles. 1. " Adam": They gave it a name. Simple Enough. 2. "A Method": They told they are presenting a method (clearing out that it's not a summary or review paper, but a new novel approach is presented here). Having this title in my mails will help me figure this out with any further info. 3. "for Stochastic Optimizations": They specify that they are providing optimization, not a new method but an improvement on current methods. 4. Done. Also, the paper itself presents Adam on the second page itself.

I know that this is not completely related to the video but do you have any advice getting letter of recommendations. I am currently an undergrad at uc berkeley studying mathematics but I find it quite hard to talk to professors other than asking questions in office hours where they explain to me. However these questions are just content of the course and I am very unknowledge about things that are beyond the course e.g. I have taken a few grad courses but all the material is very surface level to actual research where I don't know anything about. I am self studying the material for my spring semester classes this break and will hopefully try to engage with my professors more. Also another fear is that I don't want to come across trying to use them for letter of recs. I am thinking of asking them about what actual research is about.

I conduct research in botany, and one of the most striking titles I’ve come across is “From Scilla to Charybdis - Is Our Voyage Safer Now?” This work uses the metaphor of the Scylla and Charybdis legend to reflect the challenges of plant taxonomy, particularly in proposing name changes (to Charybdis) based on genetic methods. This is for me a good title! Thanks for your videos and happy 2025!

Academia is so inaccessible to the average person. First, it cost a fortune just to access articles as well to publish articles. Second, many research articles are simply published by a paper mill lab. Third, lack of research that is reproducible. Fourth, idiosyncratic language that makes it difficult for someone trying to enter the field. Academia is cooked. I do not understand why people still choose to undergo a graduate degree or especially an academic career.

Wishing you a very happy 2025!!!🎄🎄🎄See you soon! Cheers

The most awful thing are those 60 or 90 minutes "seminar talks" of PhD students... where you (not always but much too often) feel that everybody hopes for the end... heading to the lunch.

I think it is problematic if scientists state for most part of time "i have no time". What to expect from researchers who mainly have no time?

Even if your paper is cited it does not mean that it was really read. It might have been cited just for having a reasonable reference list.😅

The "too many papers which nobody reads problem" started a long time ago... I am a little bit wondering that still no paradigm change has taken place. I guess it is also a system problem... a system cannot abolish itself.

When you do R&D work, do you want the researchers to create new jobs, advance and change humanity or do you want them to just publish papers? All this depends on the governments and/or independent funding bodies themselves on how you judge the output of the researchers' work.

Dr. Churkin, thank you for sharing this insight into surviving the complexities of Academia. I am not a post-doc. I am a visitor from another planet. So please forgive the naivete of this question. In my tiny world, one's life purpose is driven by the desire to meet a perceived need. For example, if there are too many balloons in the air, there would be a need to reduce the number of balloons. And one might define one's career as "balloon-reducer," or one who addresses the problem of excess balloons on the planet. In your post-doctoral world, do researchers identify a specific need and focus the paper on solving that problem? Or is this question much too simplistic?

As someone who left academia after my PhD, I am so glad I did not stick around till my PostDoc to have this realisation. The end of your video was a bit anti-climatic for me, because you listed down exactly the type of coping mechanism one thinks off when you are still in denial of the state of academia.

Starting in 1979, if I had an idea I would develop it into a paper, send it to a magazine, later to B journals. I ended up with over 140 published papers, a handful in A Jnls. Butl the best ones were in B or C jnls. I was tenured 1983 to 2016 including 6 yrs as distinguished Prof, plus 3 phased retirement. I never worried about citations nor funding, just as well. In all professions now you have to ignore the norms and regulations, just do the job as a form on artistic expression regardless of extrinsic rewards. Somehow it works out, suprisingly. The Beatles song Hey Jude has a line: don’t you know that it’s just you,hey Jude you’ll do, the movement you need is on your shoulders.

It was helpful, thank you 😊

Hi everyone Anyone applied for PhD February 2025 intake? Have you got your ATAS ?

Very well explained, thank you

It was really helpful, thanks 😊

Does it make sense to have a 1 paper/year or 2 papers/3 years per individual, including co-authored papers? or at least some kind of limit to this effect? I'm not in this field so I'm not sure how many papers people can be expected to publish

I have a gripe with universities and publishing thesis and papers. For many tech professions, a bachelor degree is a must, for example, software engineering, but often time people in these programmes are there to get a job, and not to go into academia. I believe this might have some relation to papers becoming worthless, because people like me, who are worthless at research and papers but competent at their actual jobs, also have to write papers. This, id at least imagine, is leading to a lot of worthless papers, where people are doing the bare minimum to get a degree, thatll help getting a job. What i would suggest, if a new academic model, that has the same informational and educational value in the form of lectures, but does not require the actual writing of the thesis. For reference, i myself have a college degree and am in the last year of uni for my bachelor, and me and many of us are completely uninspired about our thesis and just want our degrees, so that we can work in peace. If we didnt have to write our papers, most of us wouldnt leaving the actually willing and motivated ones doung it, which would lessen the amount of papers being made, and potentially increasing the quality of them.

I'm sorry but your thoughts about how a corporate environment work is way off. The key deliverable is what matters and not the quality of implementation

Publish or perish. I stopped at my MS degree.

This video arguably had a greater impact than your whole PhD

You forgot to mention one important aspect of so many papers are being published everyday: they are all nonsense. It is really hard to determine what is new and what is repetition.

I was offered the opportunity for a phd in Chemistry in 2012. Back then I already felt like it's just about publishing, even in such a practical field. These days research for the heck of it is not very sustainable. The returns of investment are dimishing so you get many a disillusioned phds. I'm quite happy to have declined the phd project back then as the thing I disliked the most during my studies was writing articles/my thesis. On a positive note: Your time spent as a researcher might still have a big impact. It's just very unlikely to show in the short term. (Citation need)

You can tell someone has nothing to offer if they mention python

It seems like citations and papers published is just the easiest metric to get. If there was a way to quantify paper quality maybe we could move in a better direction

An extraordinarily well-presented and articulated presentation. The presenter is competent and articulates the content superbly. The impact is beneficial for those new to publishing manuscripts.

An exceptionally well-presented video. I enjoyed the presentation and content.

Solution? Humility

based Russian telling it like is

Optimizing for paper production is a logical way to do things when that is what is rewarded. But "max paper production" is also a local maximum in terms of performance. I am convinced that there are more important discoveries waiting out there that are down prioritized because there isn't "a clear cut paper" waiting in that work.

Hi Andrey, I liked your thoughts on the increasing number of publications from researchers. However, I would have made the title of the video as 'Declining Quality of Papers in Academia'. Research publications are and will be always valued. It's just my two cents.

Oh so that's the famous Adam Optimiser which we use in model.fit function often. 😅

An excellent explain thank you so much

On a higher level: the value of academia itself is declining. More a more resources are being put in with less and less to show for it.

Academia is basically a Soviet style planned economy where artificial metrics - publication count, impact factor - take the place of genuine price signals, which results in actors optimizing on the artificial metrics.

Another way to do research if you are in an area adjacent to business, go for an Industrial PhD instead. Much fewer positions available but you usually get a job continuing in that specific area improving it. Papers not especially relevant, they are PR and not the main reason you are hired.

The solution on the personal level: visibility. You have to publish in journals that give you visibility, publish regularly a bit above the minimum, attend the best conferences and give talks. Oh and don't forget that recognition is science comes extremely slow.

It would be auper simple to fix the metrics systems. The impact factor of a journal, the number of authors on a paper and the number of team members should be taken into account. One should simply divide the impact of a journal by the number of co-authors on a paper (if the main author does not have her/his own team). It is nonsense that the 15th author receives the same citation increase as someone who did a similar (high-ranked) paper alone. The tricky part is the number of team members. Are you a PI of big grant with 3 postdocs and 4 PhD students? Great! Your citation score should be calculated as citation/number of team members. It would take no time to develop such metrics (number of co-authors are on the papers, number of employees are on the websites), put it on a website, lean back and watch how the system is changing. Because it is very biased now and fosters to hack it by delegating the work to others and be rewarded for it.

Watching your videos really improve my presentation skills. I had a conference tomorrow and im watching this to learn how to present in front of many people. Really useful!

The supreme art of chasing after academia position (to a full professorship) is to deceive the enemy without doing any actual research work. In the practical art of academia chase, the best thing of all is to infiltrate the enemy's university academia system whole and intact; to shatter and destroy it is not so good. For one hundreds Chinese scientists to become assistant professors in one hundred universities is not the acme of skill. To deceive the enemy with research publications without doing actual scientific work is the acme of skill. If you know how the enemy's university academia system works and know yourself you need not fear the results of a hundred submitted research papers.

Your work is Gold. Thanks and keep it up. I'm an (sound) artist sneaking me in the fields of CS and your attitude and your criticism of academia is a lighthouse lit in the dark and foggy night that are the times we live on.

Great video