Neurogeek @dgmc90-blog - Tumblr Blog

The ancient Greeks argued that the best life was filled with beauty, truth, justice, play, and love. The mathematician Francis Su knows just where to find them.

#wired magazine #mathematics #math #math for everyone

Thousands of articles and tutorials have been written about data science and machine learning. Hundreds of books, courses and conferences are available. You co…

#machine learning #R #Python #data science #tutorials

In the closing weeks of 2016, Google published an article that quietly sailed under most people’s radars. Which is a shame, because it may…

Move over phrase-based language processing algorithms!

#Google #AI #Neural Machine Translation #Computation and Language #Computation #Language #Learning

Preach.

Something of concern in the scientific community and something that is becoming ever more apparent now that I’m immersed in data science is the need for more robust statistics. My first venture into the world of p-hacking was learning how to run and interpret linear mixed effects models in R, something which I continue to develop.

The article is just over a year old, but well worth a read and has some nice interactive examples of just how easy it is lull yourself into a false sense of security with p-values.

#p-hacking #statistics #data science #p-value

Optical illusions have a way of breaking the internet, and the latest visual trick looks like it’s well on its way. On Sunday afternoon, game developer Will Kerslake tweeted a picture of...

Cool example of how visual illusions are useful in helping us understand how the visual system operates.

#visual illusions #illusions #vision #vision science #visual neuroscience

Presented at the European Conference on Visual Perception (ECVP) in Barcelona last month. Very productive meeting and of course a gorgeous city!

#science #visual neuroscience #vision science #science conference #ECVP #Barcelona

VSS 2016 going down well! Lots of very good presentations and posters, made even better by the sunshine!

#tampa florida #science #vision sciences society #VSS2016 #VSS16

Tinker with a real neural network right here in your browser.

This is really, really cool.

#play with science #I'm a geek #neural networks #neuroscience #computational modelling

A rant on the pressure to publish during your PhD.

This is something that worries myself and many other PhD students whom I’ve spoken to. The pressure to get publications by the end of your PhD to ‘guarantee’ a job is enormous. IMO good science doesn’t always necessarily lead to significant results, and not everyone necessarily has the time to pump out lots of papers if what they’re doing actually takes a long time to get done properly. A PhD is supposed to be a learning and training experience: we are only just at the very start of our academic careers and we need time to learn how to ‘PROFESSIONALLY SCIENCE’ in order to become the well rounded, intuitive and EXCITED researchers that we should be. Not filled with dread that, although we spent weeks and weeks doing the right stats and collecting data in the correct way and trying to perfect the computational model you’ve been working on since Christmas (a point of personal anguish for myself...), it resulted in a non-significant outcome and will therefore go nowhere (other than into that folder called ‘noGood’ or the dreaded ‘old’ that you never look at again for fear of recurring disappointment).

I guess my bottom line is this - we’re at the start of our careers and deserve the time to develop. We should be spending time learning how to do good science - that’s how science will be pushed forward by any significant degree - rather than how to add another entry under the ‘Publications’ section of our CV. If we focus too much on publishing then all we will be good at is writing papers about bad science and that, for me, is disheartening and disappointing at the start of my academic career.

#phd student #phd life #academia #academic publishing #academic pressure

I can always get on board with rainbow unicorns …

#open science #publication #science #review process

A nice example of the usefulness of using somewhat 'abstract' stimuli to investigate midlevel vision.

#midlevel vision #visualneuroscience #vision #computationalmodelling

Most of my day has been spent trying to figure this stuff out - great post!

#mixed modelling #R #lmer #statistics

A video showing basic usage of the "lme" command (nlme library) in R. In particular, I compare output from the lm() command with that from a call to lme(). H...

Really nice introductory video by Dr. Scherber at Göttingen for building (and understanding!) mixed effects models in R.

The common approach to the multiplicity problem calls for controlling the familywise error rate (FWER). This approach, though, has faults, and we point out a fe...

It’s a fun day of statistics for me!

#statistics #hypothesis testing #multiple comparisons #p-value

Chris is good to go in the lab.

#eeg #visionscience #vision

Trending Blogs

Recently Viewed Blogs

Neurogeek