23/ Germany – a country proactively shutting down nuclear power plants despite suffering a massive energy crisis – is back to burning wood for power on an enormous scale. For the planet, of course. You could not make it up if you tried.
I’ve been using #AnnotatedEquations in my recent papers. I think it really adds to the readability and understanding of the math.
Here are some examples. It uses #tikz in #latex.
Let me know if you like it. Happy for any feedback. [GitHub link: next tweet]
#AcademicChatter
+
Use your mad neuro-modelling skills to understand un-learning in animals and humans! Cheng Lab at @ruhrunibochum still accepting applications for postdocs. Stimulating & supportive research environment, great people! Can highly recommend! https://t.co/lAB93fq7D1 @NeuroRumblr
Advances in computational neuroscience and neuromorphic engineering are offering bio-inspired alternatives to AI which perform with orders of magnitude less energy requirements, faster latency and unique advantages. Learn more from @darsalaj at #bbuzz 2021 https://t.co/f8d9ItYuys
@heikecstein @jmourabarbosa @NatureNeuro Congratulations!! I just commented on the co-author's tweet. Very exciting work! Can't wait to read your paper! https://t.co/IIN6xjhIYw
@jmourabarbosa Very exciting work! 👏 Especially for me (!) as my group and I recently submitted a very related paper where we show that Spike-frequency adaptation can be used as a powerful working memory mechanism 🧠:
https://t.co/2KhOtRplXE
@jmourabarbosa Very exciting work! 👏 Especially for me (!) as my group and I recently submitted a very related paper where we show that Spike-frequency adaptation can be used as a powerful working memory mechanism 🧠:
https://t.co/2KhOtRplXE
Pre-print of our latest work is out! It shows that spike frequency adaptation provides a surprisingly good mechanism for long short-term memory. Work done with @darsalaj, @ckraisnikovic, Guillaume Bellec, @CompLegenstein and Wolfgang Maass.
Finally, we released the code for efficient training of recurrent spiking nets:
https://t.co/N0peQ8Y6ik
This reproduces sMNIST results where spiking nets perform on the level of LSTMs.
Stay tuned for more tasks and TF2 version 🤓
@hardmaru@BellecGuill@anandsubramoney