iiinova — where your digital self keeps evolving

Joachim W Pedersen @joachimwpedersen.bsky.social · 24/02/2026

My colleagues are hosting a workshop for this years #GECCO: evolving-self-organisation-workshop.github.io/gecco-2026/ The headline is Evolving Self-Organisation. Can't wait to see all your interesting submissions! Great for #ALife and #ALICE researchers as well!

evolving-self-organisation-workshop.github.io

EvoSelf 2026

051

Reposted by Joachim W Pedersen

Sakana AI @sakanaai.bsky.social · 30/05/2025

Introducing The Darwin Gödel Machine sakana.ai/dgm The Darwin Gödel Machine is a self-improving agent that can modify its own code. Inspired by evolution, we maintain an expanding lineage of agent variants, allowing for open-ended exploration of the vast design space of such self-improving agents.

1358

Reposted by Joachim W Pedersen

Sakana AI @sakanaai.bsky.social · 12/05/2025

“Continuous Thought Machines” Blog → sakana.ai/ctm Modern AI is powerful, but it's still distinct from human-like flexible intelligence. We believe neural timing is key. Our Continuous Thought Machine is built from the ground up to use neural dynamics as a powerful representation for intelligence.

37315

Joachim W Pedersen @joachimwpedersen.bsky.social · 26/03/2025

New submission deadline: April 2nd! So still some time to put interesting thoughts on Evolving Self-Organization together! Also: We are very fortunate to have the great Risto Miikkulainen as the keynote speaker at the workshop! Can't wait to see you all there! 🤩🙌 #Evolution #Gecco #ALife

040

Joachim W Pedersen @joachimwpedersen.bsky.social · 19/03/2025

Very satisfying to see one's code run on actual real-world robots and not just simulation. Check out the paper here: arxiv.org/pdf/2503.12406

arxiv.org

040

Joachim W Pedersen @joachimwpedersen.bsky.social · 19/03/2025

www.youtube.com/watch?v=jnoa... Bio-Inspired Plastic Neural Nets that continually adapt their own synaptic strengths can make for extremely robust locomotion policies! Trained exclusively in simulation, the plastic networks transfer easily to the real world, even under various extra OOD situations.

youtube.com

[IROS25] Bio-Inspired Plastic Neural Nets for Zero-Shot Out-of-Distribution Generalization in Robots

YouTube video by Worasuchad Haomachai

2205

Joachim W Pedersen @joachimwpedersen.bsky.social · 26/02/2025

Remember that 4-page submissions of early results are also welcome! Also, does anyone know if #GECCO has an official 🦋 account? I cannot seem to find it...

010

Joachim W Pedersen @joachimwpedersen.bsky.social · 10/02/2025

Both 4-pagers of early research as well as 8-page papers with more substantial results are welcome!

010

Joachim W Pedersen @joachimwpedersen.bsky.social · 10/02/2025

Join us for the Evolving Self-Organisation workshop at #GECCO this year! Great chance to submit your favourite ideas concerning self-organisation processes and evolution, and how they interact. Relevant for Alifers #ALife and anyone interested in #evolution, #self-organisation, and #ComplexSystems.

1143

Joachim W Pedersen @joachimwpedersen.bsky.social · 21/01/2025

Very cool! And great aesthetics as well 🙌 😊

030

Reposted by Joachim W Pedersen

Sebastian Risi @risi.bsky.social · 21/01/2025

Ever wish you could coordinate thousands of units in games such as StarCraft through natural language alone? We are excited to present our HIVE approach, a framework and benchmark for LLM-driven multi-agent control.

23811

Joachim W Pedersen @joachimwpedersen.bsky.social · 15/01/2025

With all the research coming from Sakana AI, this figure needs to be updated fast! direct.mit.edu/isal/proceed... #LLM #ALife #ArtificialIntelligence

0131

Reposted by Joachim W Pedersen

hardmaru @hardmaru.bsky.social · 15/01/2025

Transformer²: Self-adaptive LLMs arxiv.org/abs/2501.06252 Check out the new paper from Sakana AI (@sakanaai.bsky.social) paper. We show the power of an LLM that can self-adapt its weights to its environment!

14610

Reposted by Joachim W Pedersen

IT-Universitetet i København @itu.dk · 14/01/2025

Vi har samlet et starter pack med forskere og repræsentanter fra ITU på Bluesky. Mød dem her 👇 go.bsky.app/E8WJwXS

0249

Reposted by Joachim W Pedersen

Lazaros Nalpantidis @lanalpa.bsky.social · 08/01/2025

Can Dynamic Neural Networks boost Computer Vision and Sensor Fusion? We are very happy to share this awesome collection of papers on the topic!

062

Reposted by Joachim W Pedersen

Mattia Rigotti @matrig.net · 22/12/2024

If microchip ~= silicon then AGI ~= huge pile of sand

1192

Reposted by Joachim W Pedersen

hardmaru @hardmaru.bsky.social · 10/12/2024

Neural Attention Memory Models are evolved to optimize the performance of Transformers by actively pruning the KV cache memory. Surprisingly, we find that NAMMs are able to zero-shot transfer its performance gains across architectures, input modalities and even task domains! arxiv.org/abs/2410.13166

1579

Joachim W Pedersen @joachimwpedersen.bsky.social · 04/12/2024

3) Optimizer optimization: Think hyperparameter tuning, e.g., learning rate etc. The search within the inner-loop is altered. We use meta-learning to achieve improved inner-loop optimization, so it is well worth considering exactly how our double-loop achieves this! #meta-learning #deeplearning #ai

020

Joachim W Pedersen @joachimwpedersen.bsky.social · 04/12/2024

1) Starting point optimization: Think MAML. Move the initial point of the inner-loop search to a better place to learn quick. 2) Loss landscape optimization: Think neural architecture search. The loss landscape(s) of the inner-loop is transformed.

110

Joachim W Pedersen @joachimwpedersen.bsky.social · 04/12/2024

This can be thought of independently from which optimizer is being used in the inner-loop. In any meta-learning approach, the outer-loop optimization will transform the inner-loop optimization process in at least of one three ways and often in a combination of these three.

110

Joachim W Pedersen @joachimwpedersen.bsky.social · 04/12/2024

In deep learning research, we often categorize meta-learning approaches as either gradient-based or black-box meta-learning. In my PhD thesis, I argued that it can sometimes be useful to classify approaches based on how the outer-loop optimization affects the inner-loop optimization.

162

Reposted by Joachim W Pedersen

Craig Reynolds @craigreynolds.bsky.social · 28/11/2024

Like 130,000 others, I made a starter pack. This one is people working on or with evolutionary computation in its many forms: genetic algorithms, genetic programming, evolution strategies. If you like to be added, or suggest someone else, message me or reply to this post.

74012

Joachim W Pedersen @joachimwpedersen.bsky.social · 28/11/2024

Thanks for making a pack putting the spotlight on evolutionary computation! I would love to join the list :)

110

Joachim W Pedersen @joachimwpedersen.bsky.social · 24/11/2024

Would also like to plug this perspective paper: From Text to Life: On the Reciprocal Relationship between Artificial Life and Large Language Models, which I was proud to contribute to along with all the talented colleagues on the author list :)

110

Joachim W Pedersen @joachimwpedersen.bsky.social · 24/11/2024

In the same vein of NDPs: arxiv.org/pdf/2405.08510 A have also been interested in synaptic plasticity for a while ( arxiv.org/pdf/2104.07959 ) and how plasticity can be used to achieve structural flexibility in neural networks: dl.acm.org/doi/pdf/10.1...

arxiv.org

120

Joachim W Pedersen @joachimwpedersen.bsky.social · 24/11/2024

Hi, I'm Joachim, I work as a postdoc on the Grow-AI project lead by @risi.bsky.social at ITU of CPH. My PhD is titled "Bio-Inspired Approaches to Adaptive Artificial Agents" and since then, a lot of my time has revolved around Neural Developmental Programs: watermark.silverchair.com/isal_a_00755...

watermark.silverchair.com

161

Reposted by Joachim W Pedersen

Antonino Greco @agreco.bsky.social · 17/10/2024

For the Blueskyers interested in #NeuroAI 🧠🤖, I created a starter pack! Please comment on this if you are not on the list and working in this field 🙂 go.bsky.app/CscFTAr

11012050

Joachim W Pedersen @joachimwpedersen.bsky.social · 23/11/2024

Hi, I'd love to join the list as well:)

110

Joachim W Pedersen @joachimwpedersen.bsky.social · 17/11/2024

Thanks! Culture is definitely an accelerator, but still leaves some questions: Isn't the ability to outsmart adversaries always adaptive whether they are part of your species or not? And if culture is what made humans so smart, why are sophisticated cultures not more widespread in evolution?

010

Reposted by Joachim W Pedersen

Manlio De Domenico @manlius.bsky.social · 16/11/2024

A visionary Ross Ashby writing this, 62 years ago: “A type of system that deserves much more thorough investigation is the large system that is built of parts that have many states of equilibrium” csis.pace.edu/~marchese/CS... #ComplexSystems #Cybernetics

24917

Joachim W Pedersen @joachimwpedersen.bsky.social · 17/11/2024

So is high intelligence with a wide capacity of learning and abstracting only worth it in some niches and not generally adaptive? What do you all think? And what are some good resources to learn more about these arguments? n/n

000

Joachim W Pedersen @joachimwpedersen.bsky.social · 17/11/2024

Counter point to that is that even today, in the world of insects there are species that are arguably as successful as humans, and there is currently nothing that suggests that e.g. ants or termites are on a trajectory to abstract thought.

100

Joachim W Pedersen @joachimwpedersen.bsky.social · 17/11/2024

Another point could be that on a longer time span, high intelligence will actually be a dominating trait for 99 % of time. If it is a trait that is here to stay, then in a few billion years the era without abstract thought could seem like a small blip.

100

Joachim W Pedersen @joachimwpedersen.bsky.social · 17/11/2024

Obviously more complexity takes longer time to evolve, larger brains require more energy, etc. But pheno-plasticity is known to smoothen the loss landscape and brain anatomy across high and low intelligence species are similar enough to suggest they are not far from each other in the genome space.

100

Reposted by Joachim W Pedersen

Manuel Baltieri @manuelbaltieri.bsky.social · 13/11/2024

Here's an attempt to bring together ALife researchers and enthusiasts here, who am I missing? go.bsky.app/LgoJN2N

16269

Joachim W Pedersen @joachimwpedersen.bsky.social · 14/11/2024

Thanks for making this list! I would love to join as well:)

000

Joachim W Pedersen @joachimwpedersen.bsky.social · 14/11/2024

Nothing like a country-side writing retreat in cozy surroundings! Expect new things from the Grow-AI project soon!

030