Jonathan Shock (@jonsadventures) profile picture

Jonathan Shock

@jonsadventures

Associate Professor at the University of Cape Town

 Cape Town, South Africa
shocklab.net
Joined February 2009
  • Tweets100
  • Following191
  • Followers223
9Media
Photo by Jonathan Shock (@jonsadventures)Photo by Jonathan Shock (@jonsadventures)Photo by Jonathan Shock (@jonsadventures)Photo by Jonathan Shock (@jonsadventures)Photo by Jonathan Shock (@jonsadventures)Photo by Jonathan Shock (@jonsadventures)Photo by Jonathan Shock (@jonsadventures)
If you are interested in doing a #PhD with me at Imperial College London and qualify as a home student, please reach out (before end of 2024)! Potential topics: spatial statistics, deep generative models, probabilistic programming and more.
18
1
27
3,210
Even More Generalized Hamiltonian Dynamics. arxiv.org/abs/2408.14420
1
1
150
W. A. Horowitz, A. Rothkopf: Even More Generalized Hamiltonian Dynamics arxiv.org/abs/2408.14420 arxiv.org/pdf/2408.14420
1
1
181
Excited to announce Jumanji v1.0, now featuring 22 fast, flexible, and scalable environments! Fully written in JAX, Jumanji offers on-device fully-jitted simulations and training. Jumanji got published at ICLR 2024! Paper: arxiv.org/abs/2306.09884 GitHub: github.com/instadeepai/juman…
1
27
3
124
31,247
Title: Growing the MARL software ecosystem in #JAXSpeakers: MARL Research Team @instadeepai Wednesday, 20 March 2024, 16h00 (GMT +2) zoomto.me/FwF7KThis crew have been doing amazing work! Please share with anyone who may be interested.
Image from Jonathan Shock (@jonsadventures)
3
11
425
I am extremely grateful to my co-organisers @UT3PaulSabatier @IFAS_Research, and to the speakers and the delegates from the @FrenchEmbassyZA @NRF_News @WitsUniversity and @CNRSinSthAfrica for their contribution and support to the #BrAInEvolution conference. We'll share more soon!
Image from Amélie Beaudet (@BeaudetAmelie)
9
39
2,267
It has been a productive and exciting week for the #EndoMap team in South Africa! Among other things, we had the privilege to spend some time in the Cradle of Humankind and with the fossils, meet with international colleagues and brainstorm on the future of paleoneuroscience🧠
Image from Amélie Beaudet (@BeaudetAmelie)
Image from Amélie Beaudet (@BeaudetAmelie)
Image from Amélie Beaudet (@BeaudetAmelie)
5
22
1,706
Jonathan Shock retweeted
Mark Solms (@Mark_Solms) profile picture
Mark Solms@Mark_Solms22 Nov 2023
I am very proud of these wines. My (branch of my) family's main activity for the past 500 hundred years has been wine making. However, please remember that since I took over our South African estate in 2001, ALL the profits have gone to the farm workers.
9
4
1
94
11,902
mathemafrica.org/?p=17167 on Kevin Mitchell's book Free Agents - How Evolution Gave us Free Will.
1
113
With some of my current and former Masters and PhD students at the #Indaba2023
Image from Jonathan Shock (@jonsadventures)
3
184
I have six of my Masters and PhD students here at the #indaba2023 which has been phenomenal so far. Over 700 ML researchers and practitioners from around Africa. My poster was brilliantly put together by Alyssa Amod and Kaitlyn Sparks. tinyurl.com/26vs28tw
Image from Jonathan Shock (@jonsadventures)
1
3
289
Photos are of my Masters and PhD students, and one honours student. The names around the outside are collaborators and collaborating groups.
86
'Sophisticated Learning' (SL) is a new extension of 'sophisticated active inference' (SI) planning algorithms. SL combines SI with elements of Bayesian RL and outperforms both on specific planning problems. arxiv.org/abs/2308.08029 Developed by @rowan_hodson. Feedback welcome!
Sophisticated Learning: A novel algorithm for active learning...

Sophisticated Learning: A novel algorithm for active learning...

We introduce Sophisticated Learning (SL), a planning-to-learn algorithm that embeds active parameter learning inside the Sophisticated Inference (SI) tree-search framework of Active Inference....

arxiv.org
1
13
39
4,662
I just did a cool experiment using GPT to unpack some Active Inference & FEP formalisms, based on the equations for our recent paper: activeinference.org/research… arxiv.org/abs/2303.04898 For this paper we made LaTeX representation of all equations, which renders nicely in Coda. coda.io/@active-inference-in… Then I used GPT-4 to explain the equation from the LaTeX, this is equation 2 from the paper. See the images for the prompt and response.
Image from Active Inference Institute (@InferenceActive)
Image from Active Inference Institute (@InferenceActive)
Image from Active Inference Institute (@InferenceActive)
Image from Active Inference Institute (@InferenceActive)
1
5
20
2,303
The Mind and Matter conference in Helsinki has been an absolutely wonderful couple of days of deeply stimulating discussion and talks. helsinki.fi/en/conferences/m…
Programme | Mind and Matter 2023 | University of Helsinki

Programme | Mind and Matter 2023 | University of Helsinki

Sessions 1-5 take place in Metsätalo, room 1 (Unioninkatu 40). The reception and the poster session are held in the banquet rooms at Unioninkatu 33.

helsinki.fi
1
137
I gave a couple of talks recently for the general public on an introduction to Artificial Intelligence, to the Oxford University IT Learning Centre, thanks to @ITLCOxford @EmmaProcterLegg . Here's a recording of the first: youtube.com/watch?v=vGgxnvyW…

AI: A lightning tour of AI

Professor Jonathan Shock introduces us to the world of modern AI, g...

youtube.com
1
1
2
308
OUT NOW: pre-print non-final PDF of our book "Multi-Agent Reinforcement Learning: Foundations and Modern Approaches" released on official page marl-book.com. The authors (I, @f_christianos, @LukasSchaefer96) will be at @aamas2023 & @ieee_ras_icra in London. @mitpress
Multi-Agent Reinforcement Learning: Foundations and Modern Approaches

Multi-Agent Reinforcement Learning: Foundations and Modern Approaches

The first comprehensive textbook on multi-agent reinforcement learning. MIT Press, 2024. Free PDF, algorithm code and lecture slides.

marl-book.com
Stefano V. Albrecht (@s_albrecht) profile pictureStefano V. Albrecht@s_albrecht8 May 2023
(1/2) Update: our MARL book is on track for printing with @mitpress in 2024! We will release the pre-print book PDF in Q3 this year. 📔 As a teaser, here is the current book ToC (may change in final version) + GPT version below. More details in original announcement below.
Image from Stefano V. Albrecht (@s_albrecht)
Image from Stefano V. Albrecht (@s_albrecht)
Image from Stefano V. Albrecht (@s_albrecht)
7
37
1
118
38,696
I found a series of lectures on YouTube of Paul Dirac from 1975, talking about QM. The audio quality was extremely low, so I have cleaned them up using Adobe Podcast Enhance and iMovie background noise remover. Here are the enhanced videos: youtube.com/watch?v=2GwctBld…
Dirac lecture 1 of 4 - Quantum Mechanics - very clean audio

Dirac lecture 1 of 4 - Quantum Mechanics - very clean audio

This is a video of Dirac's first lecture of four on quantum mechani...

youtube.com
1
3
159
@markcwebster thank you for the amazing product!
74
🎉Join us at the 1st Annual Web Conference of the International Society for the Philosophy of the Sciences of the Mind! No need to worry about carbon footprint, travel expenses, or dressing up - just bring your quirky ideas and join us online!
Neural Mechanisms Online (@NeuralMech) profile pictureNeural Mechanisms Online@NeuralMech14 Mar 2023
CfP 1st Annual Conference of the International Society for the Philosophy of the Sciences of the Mind ispsmind.com/international-c… main info: Paper submission by May 21st Conference: 24-25 Nov and 1-2 Dec Venue: online Best meme prize‼️
Image from Neural Mechanisms Online (@NeuralMech)
4
17
3
87
17,107
Load more