Happy to share a preprint to the paper outlining the work I took part in as part of an internship at @merckgroup last year. We present an artificial neural network capable of solving a wide variety of cognitive tasks framed as graph traversal problems #AI#neuroscience
Preprint final paper headlined „Can the brain use waves to solve planning problems?” by Inst Neurosci.+Psychol. @UofGlasgow and us (with help from @TINSInstitute@TINSNeurotwin) about understanding the #computational principles of human brain. #Science
https://t.co/H6XbJFH8B7
Awesome video of the revolutionary techniques of Nikolai Bernstein to study human kinematics. The questions outlined in his book 'The Coordination and Regulation of Movements' are still at the forefront of motor control research today.
https://t.co/p1KciVXE8T
@katie_riddoch@MorpheusFaustin@PhDVoice@AcademicChatter@TweeterAcademic If the kinds of ML you're doing don't require this then a gaming laptop is not necessary and a Mac is probably more advisable as it will be considerably easier to set up and work with. Hope that helps and feel free to send me a DM if you need any more help.
@katie_riddoch@MorpheusFaustin@PhDVoice@AcademicChatter@TweeterAcademic Basically I would wait until you have an idea of exactly what kind of computing power you need. Most universities have high performance computing clusters that you can use for free and which will be orders of magnitude faster than anything you can buy commercially.
@katie_riddoch@MorpheusFaustin@PhDVoice@AcademicChatter@TweeterAcademic I just realized you may not have access to a HPC so in that case: If you are doing deep learning or any kind of ML that involves large amounts of matrix algebra then you'll need a laptop with a good GPU (gaming laptops are best for this). 1/2
@katie_riddoch@MorpheusFaustin@PhDVoice@AcademicChatter@TweeterAcademic I bought an expensive gaming laptop at the start of my PhD to train deep learning models and my main take away had been: it is not at all necessary to spend lots of money on a laptop in order to be able to do ML research. 1/2
@UofGlasgow and @UniWestScotland want to better understand the human-pet bond. Help us by completing this 10 minute survey, and sending a video of your dog! Participants will be entered into a prize draw for one of 5 £20 Amazon Gift Cards. #dogsofinstagram https://t.co/3ByWeeFZeV
Curtis Flowers was tried 6 times for the same crime, which creates a rare kind of evidence for the effect of jury composition on verdicts.
https://t.co/DqSgOomeCG
The probability that an all-white jury would convict is 55%.
A "fair cross-section" jury, only 15%. @innocence
Losing my absolute mind at this documentary hosted by Pierce Brosnan about the legend of Robin Hood. It is one of the strongest accidental Partridges you will ever witness