Pierre-Yves Oudeyer @pyoudeyer Twitter profile

Pinned Tweet

Pierre-Yves Oudeyer

4 years

We're thrilled to inaugurate our new blog @FlowersINRIA with this post: Language as a Cognitive Tool: Dall-E, Humans and Vygotskian RL Agents Thanks @cedcolas @karch_tristan for the outstanding work! 1/21

1

37

129

Last Seen Profiles

@stw46

@Amequ2

@midoelars

@TheDeepeshJJN

@ahmedmona111111

@gasmanthegascan

@YozoHoshizora

@0xCaptainLevi

@BinorRaja

@dertgdssrtc

@T_HighAthletics

@stw46

@docentdemagogue

@ttymbell

@MacAnsari

@y_futatsugi

@Goitsione_dee

@Cstarz305

@Pat1e2

@stwmaniax

@CarryMcBarry

@StephaneP1295

@CLAYTONCARVALH3

@synyx_ka

@Palmefotografia

@cawtoru

@d1x1ee7

@nataliesxn12

@CrazyThumbzz

@SaaraOsterberg

@stwmaniax

@koosamagoopaloo

@nataliesxn12

@badorcc2019

Pierre-Yves Oudeyer

@pyoudeyer

11 months

Take Flow Lenia, add random inflow and outflow of matter + ecological niches, what happens ? 🧐🔬 💡Spontaneous self-organization of an evolution process ? Generating exploring "agents"?

hardmaru

@hardmaru

11 months

Artificial lifeforms are super fascinating to watch. These self-organizing, self-replicating, “lifeforms” emerged from a continuous time cellular automata system called Flow-Lenia. Lenia is a family of CAs generalizing Conway’s Game of Life to continuous space, time and states.

33

526

3K

13

126

546

Pierre-Yves Oudeyer

@pyoudeyer

6 years

How many random seeds are needed to compare #DeepRL algorithms? Our new tutorial to address this key issue of #reproducibility in #reinforcementlearning PDF: Code: Blog: #machinelearning #neuralnetworks

How Many Random Seeds Should I Use? Statistical Power Analysis in (Deep) Reinforcement Learning...

Reproducibility in Machine Learning and Deep Reinforcement Learning in particular has become a serious issue in the recent years. Reproducing an RL paper can turn out to be much more complicated than...

openlab-flowers.inria.fr

5

174

416

Pierre-Yves Oudeyer

@pyoudeyer

6 years

Our new #Corl18 paper: How curiosity-driven autonomous goal setting enables to discover independantly controllable features of environments pdf: Blog: Colab: #machinelearning #AI #NeuralNetworks

Curiosity_Driven_Goal_Exploration.ipynb

Colaboratory notebook

colab.research.google.com

6

130

305

Pierre-Yves Oudeyer

@pyoudeyer

6 years

Interesting paper and review on meta-learning: #deeplearning #neuralnetworks

2

77

258

Pierre-Yves Oudeyer

@pyoudeyer

3 years

Super happy to announce I joined Microsoft Research Montreal for a 10 months research visit! I'll explore new projects at the frontiers of #AI #DeepRL #language in an amazing team and ecosystem! 😀🚀 Thanks @MSFTResearch @susan_dumais @harmvanseijen and all for the opportunity🙏

5

3

220

Pierre-Yves Oudeyer

@pyoudeyer

1 year

Can #LLMs help children learn in classrooms? 🧒🏼🎓 💡Yes! Super happy to share this field study wth #GPT helping 9-10 yo to learn asking curiosity-driven questions!🔍🤔 Amazing work led by @RaniaABDBenAmar @FlowersINRIA @T_EvidenceB #education #edTech 📖

8

78

218

Pierre-Yves Oudeyer

@pyoudeyer

2 years

New conference on Curiosity, Creativity and Complexity in humans and machines ! With very interdisciplinary panel of speakers in neuroscience, psychology, economics, AI, ... 👉 May 23-25th, NY, information to submit abstracts or just attend:

2

59

204

Pierre-Yves Oudeyer

@pyoudeyer

1 year

A great recent news in the team @FlowersINRIA was the Best Paper award at Alife conference🏅 It introduced Flow Lenia, a CA wth mass conserv. & local evolving rules, opening possibilities for self-organization of evolution processes 📺 @eplantec 's talk is now available !!! 🎉👇

erwan plantec

@eplantec

1 year

Video of my presentation at ALIFE 2023 is now available ! I present Flow Lenia, a mass conservative CA in which life reminiscent creatures emerge, interact and potentially evolve ! 📄 )(🧵 )

1

14

91

9

55

199

Pierre-Yves Oudeyer

@pyoudeyer

3 years

In the beginning, there is only an environment. No body, no sensors, no agents. How could creatures self-organize? Able to move, adapt, interact, generalize? Use gradient descent + curriculum learning! 👉 New blog post: 1/n

Learning Sensorimotor Capabilities in Cellular Automata

See blog post and interactive demo atdevelopmentalsystems.org/sensorimotor-lenia/

www.youtube.com

3

46

185

Pierre-Yves Oudeyer

@pyoudeyer

5 years

"Education of Machinery" What a wonderful name for #machinelearning , in Turing 1948 paper on learning #neuralnets

hardmaru

@hardmaru

5 years

Alan Turing proposed in 1948 the use evolution to “train” a particular type of neural network he called a “B-type unorganised machine”. He evolved binary networks to perform all sorts of tasks back then.

16

329

1K

1

54

175

Pierre-Yves Oudeyer

@pyoudeyer

10 months

🌍Foundational models are cultural technologies They (will) mediate and influence human culture in ways that are unprecedented: here are the real opportunities 🚀and challenges🚧of #AI today ! 📚Here's a paper to make progress understanding them 💡👇

Levin Brinkmann

@LevinBrinkmann

10 months

New Paper: "Machine Culture" AI systems are increasingly integrated into the human cultural fabric. What can we learn from human cultural evolution to anticipate the impact of intelligent machines on culture (socially transmitted information)? [1/9]

2

60

161

3

41

171

Pierre-Yves Oudeyer

@pyoudeyer

11 months

🌐Open-ended learning may be the new #AI frontier It's about enabling machines to learn after the've already learned all existing data 🤖 📚 ❓ But what's "open-ended learning"? 🤔 📖 Here are steps towards better understanding this concept:

2

43

160

Pierre-Yves Oudeyer

@pyoudeyer

5 years

Automatic Curriculum Learning For Deep RL: A Short Survey What is ACL? Why use ACL? Which forms of ACL (e.g. goal generation, reward shaping, procedural gen. of envs, self-play)? What open questions? w @RPortelas @cedcolas @lilianweng @katjahofmann

2

50

157

Pierre-Yves Oudeyer

@pyoudeyer

7 years

I just posted a new review on computational theories of curiosity-driven #learning where I discuss how tools from #MachineLearning and #devrob help us understand better human #curiosity #ArtificialIntelligence #AI @cogdevsoc #deepRL

2

49

153

Pierre-Yves Oudeyer

@pyoudeyer

4 years

Huge congrats @mayalen_etc for your amazing work on this #Neurips2020 paper (oral presentation!): Meta-diversity search + learning of hierarchies of representations for automated discovery of novel structures in continuous #cellularautomata

4

30

148

Pierre-Yves Oudeyer

@pyoudeyer

6 years

I am highly honored to announce I received the 2018 @inria prize from French National Academy of Science (researcher under 40) for my work in #AI #machinelearning #cognitivescience esp. on the topic of #curiosity driven learning

14

136

Pierre-Yves Oudeyer

@pyoudeyer

4 years

Super happy to share this survey within the field of developmental #AI we've helped to grow in the last decade!!! 👉 Intrinsically motivated goal-conditioned reinforcement learning Special thanks @cedcolas @karch_tristan for the outstanding work! 1/17

1

32

131

Pierre-Yves Oudeyer

@pyoudeyer

6 years

Our new @icmlconf paper is out! We show how population-based diversity search can bootstrap #deepRL for hard exploration problem PDF: Blog: Code: #NeuralNetworks #deeplearning #AI

1

47

129

Pierre-Yves Oudeyer

@pyoudeyer

6 years

Today with @forestier_seb we made a fun discovery while setting up a live robot learning experiment for a tv documentary project. We aimed to compare runs of curiosity-driven goal exploration and a RL scheme (sparsely) rewarded for moving a ball

1

36

123

Pierre-Yves Oudeyer

@pyoudeyer

6 years

« algorithmic information theory essentially says that the probability of producing some types of outputs is far greater when randomness operates at the level of the program describing it rather than at the level of the output itself »

0

24

113

Pierre-Yves Oudeyer

@pyoudeyer

6 years

Highly honored to have the opportunity to give an invited talk @iclr2019 ! I will talk about developmental autonomous learning, intrinsic motivation and autonomous goal setting from 4 interrelated perspectives: #AI , #developmentalrobotics , #cognitivescience and #edTech @inria

ICLR 2025

@iclr_conf

6 years

Excited to preview our list of invited speakers for #iclr2019 in New Orleans: Léon Bottou, Cynthia Dwork, Ian Goodfellow, Noah Goodman, Mirella Lapata, Pierre-Yves Oudeyer, Emily Shuckburgh, and, Zeynep Tufekci Tickets go on sale tomorrow @ 4pm PST at

3

35

202

2

11

108

Pierre-Yves Oudeyer

@pyoudeyer

4 years

Super happy that the #popularscience book for kids I wrote with @dphroy is out! It introduces #AI , #machinelearning and #robotics to >7 years old (in french :)) It was a lot of fun to conceive & now rewarding to see in the hands of children! 1/n

6

32

106

Pierre-Yves Oudeyer

@pyoudeyer

3 years

Introducing TeachMyAgent, a Benchmark for Automatic Curriculum Learning in #DeepRL @ #ICML2021 ! PCG environments + #DeepRL learners + diverse robot morphologies + many tests + 7 baselines Web: Paper: Code:

2

17

105

Pierre-Yves Oudeyer

@pyoudeyer

7 years

The Lighthill debate (1973): Extraordinary video to understand the history of #ArtificialIntelligence and how winter came. Many arguments on both sides are still valid today about #deeplearning approaches @ylecun @GaryMarcus @fchollet @hardmaru #AI

The Lighthill debate on Artificial Intelligence: "The general purpose...

Historical video about artificial intelligence and how the AI winter came in the 1970sThis is the full video of the Lighthill debate on Artificial Intelligen...

www.youtube.com

3

55

92

Pierre-Yves Oudeyer

@pyoudeyer

5 years

Honored to receive the Atos - Joseph Fourier prize for work on curiosity-driven #AI and its applications ! A big thank to the many great colleagues/students I have been working with on this topic over the past decade !

Cyril Giraudeau

@CyrilGiraudeau

5 years

Rendre l' #iA autonome dans son apprentissage comme un enfant le ferait, pour rendre cette phase bien moins consommatrice d'énergie, c'est que projet de @pyoudeyer récompensé par le prix @AtosFR Fourier 2019 Félicitations @Inria Bordeaux !

0

6

22

6

15

98

Pierre-Yves Oudeyer

@pyoudeyer

6 years

Great slide from Alex Graves at the unsupervised #deeplearning tutorial, outlining a 2-dimensional organization of learning approaches #NeurIPS2018

1

27

101

Pierre-Yves Oudeyer

@pyoudeyer

2 years

Dans les collèges et lycées tout le monde parle de #ChatGPT 🦜🧑‍🎓🗣️ Mais combien savent comment ça fonctionne ?⚙️ Quelles sont ses possibilités et ses limites ? 🤔 Pour les enseignants et les élèves, voici un tour d'horizon en 5 mn 🚀🤖💡 #education

Comment fonctionne ChatGPT? Un tour d’horizon en moins de 5mn

Cette vidéo est une introduction aux modèles de langage, qui sont à la base d’outils comme #chatgpt ou Bard. Elle s'adresse à un public large (par exemple él...

www.youtube.com

2

74

99

Pierre-Yves Oudeyer

@pyoudeyer

3 years

New blog post from @RPortelas ! Teacher Algorithms for #DeepRL Agents that Generalize in Procedurally Generated Environments Presents the emerging Deep RL sub-field of automatic curriculum learning with a case-study on ALP-GMM

0

25

98

Pierre-Yves Oudeyer

@pyoudeyer

1 year

Many practitioners observe that pretraining LLMs on code may boost natural language "reasoning" abilities Do you know papers actually showing this through systematic scientific experiments ? @DrJimFan @Francis_YAO_ @gneubig @srush_nlp @Thom_Wolf

10

16

96

Pierre-Yves Oudeyer

@pyoudeyer

4 years

You want to build a fun in-browser interactive demo of #DeepRL with #Box2D walkers/swimmers/climbers #javascript #tensorflow .js ? You search an internship in 2021 (master level) ? Here is an opportunity to join @FlowersINRIA

4

29

92

Pierre-Yves Oudeyer

@pyoudeyer

4 years

How can an artificial scientist explore the #gameoflife to discover a diversity of novel creatures and learn to represent them? That's our #ICLR2020 paper: Blog+code: wth @ScienceReinke and @mayalen_etc

1

26

81

Pierre-Yves Oudeyer

@pyoudeyer

2 years

Great to see automatic curriculum learning, the TeachMyAgent benchmark and interactive web demo included in @huggingface 's #DeepRL course !!! 🤗

Pierre-Yves Oudeyer

@pyoudeyer

2 years

2) wth @ClementRomac , development of the TeachMyAgent benchmark for systematic comparison of automatic curriculum learning algorithms in #DeepRL (= teacher algorithms) 👉

1

0

3

1

19

83

Pierre-Yves Oudeyer

@pyoudeyer

1 year

🧐Curiosity-driven learning has many different forms, functionalities and usages in 🐵animals & 🤖machines ▶️ We review them in this paper led by Alexander Ten and @Clement_MF_

1

24

81

Pierre-Yves Oudeyer

@pyoudeyer

5 years

Amazing: every week I see a #DeepRL paper comparing algorithms using mean performance over *3* seeds ! Yes, ****3**** !!! Please please #DeepRL community, your great ideas will be served better using standard scientific methods!

5

10

78

Pierre-Yves Oudeyer

@pyoudeyer

2 years

LLMs often encode wrong information about the world. How to mitigate this? Not by passive reading: they need to interact wth the env, make causal intervention, get feedback & update. To understand better this process, systematic studies in simple envs are needed, here's one👇

Clément ROMAC

@ClementRomac

2 years

🚨🚨 Large Language Models encode world knowledge that can benefit RL agents when solving decision-making problems (e.g. in robotics 🤖). But what happens when their knowledge are wrong? 🎯 Can we align LLMs and ground their knowledge through interactions with online RL? 🕹

3

44

230

2

16

74

Pierre-Yves Oudeyer

@pyoudeyer

6 years

Interested by a #PhD on curiosity-driven multi-goal #DeepRL in #Minecraft / Malmo ? Project @FlowersINRIA in collaboration with @MSFTResearch and @katjahofmann Full description: #AI #neuralnetworks #ICLR2018

0

32

68

Pierre-Yves Oudeyer

@pyoudeyer

3 years

In this new blog post, I present my perspective on the future of #AI and an approach to scale to human-like intelligence: focus on socio-cultural *environments* 👉 1/n

1

19

73

Pierre-Yves Oudeyer

@pyoudeyer

2 years

Today @BertChakovsky is giving a talk @FlowersINRIA @inria on evolution and self-organization of intelligent agents from the ground up (= from envs where there is not already things like « bodies », « senses », « muscles », « brains » or « learning »)

2

15

75

Pierre-Yves Oudeyer

@pyoudeyer

11 months

Voici la 5ème vidéo de notre série pédagogique sur #ChatGPT pour les collégiens, lycéens, profs, parents 🦜🧑‍🎓🗣️ #IA Ici nous expliquons comment il est possible de simuler un "raisonnement" ⚙️🤔avec ChatGPT, et comment celui-ci peut dérailler 🤖😅🤯

ChatGPT en 5mn: Prompting avancé, raisonnements par chaînes de pensée...

Les modèles de langage comme ChatGPT ou Bard sont capables de tenir des raisonnements simples pour résoudre un problème donné. Cependant, il est parfois néce...

www.youtube.com

2

50

83

Pierre-Yves Oudeyer

@pyoudeyer

4 years

We're so happy to start a new collaboration with the amazing @BertChakovsky at @FlowersINRIA !!! + We open a master level internship for 2021 to study the emergence of sensorimotor behavior in #Lenia creatures such as the ones below 👾👇 #Alife #AI

Bert Chan

@BertChakovsky

4 years

The first ever continuous glider gun is... a bit messy 🥴

12

78

568

6

13

62

Pierre-Yves Oudeyer

@pyoudeyer

4 years

Here's my most up-to-date talk, with historical notes and recent new directions in developmental #machinelearning , #curiosity and #DeepRL + a focus on autonomous goal imagination & language Huge thanks @ferranAlet and the @MIT_CSAIL for invitation! 1/n

MIT EI Seminar - Pierre-Yves Oudeyer - Developmental Machine Learni...

MIT Embodied Intelligence Seminar - April 22, 2020Speaker: Pierre-Yves OudeyerTitle: Developmental Machine Learning, Curiosity and Deep RLAbstract: Current ...

www.youtube.com

1

19

67

Pierre-Yves Oudeyer

@pyoudeyer

6 years

Super cool and inspiring talk from @hardmaru on interactive exploration of generative models

Eyeo 2018 - David Ha

LEARNING ABSTRACTION WITH NEURAL NETWORKS | David Ha at Eyeo 2018 | David is a Research Scientist at Google Brain. His research interests include Recurrent Neural…

vimeo.com

1

11

65

Pierre-Yves Oudeyer

@pyoudeyer

2 years

What are the opportunities and challenges of #LLMs in #education ? 👉 A useful review and discussion from @EnkelejdaKasne1 et al.

3

17

67

Pierre-Yves Oudeyer

@pyoudeyer

2 years

A few years ago, we wrote a paper that ended up being foundational in the team @FlowersINRIA : it sparked most of our recent research on autotelic #AI and automatic curriculum learning 👉 An extended version just appeared in the @JmlrOrg journal !!! 🚀🥳 📰

1

7

66

Pierre-Yves Oudeyer

@pyoudeyer

1 year

The future of #AI may not be generalist #LLMs , i.e. #AGI Rather it may be 'societies' of diverse specialized models In this thread, I'll discuss GPT-42 (yes, 42 :)) and links with cultural evolution, open-source, industry and open-endedness 🧵

2

19

65

Pierre-Yves Oudeyer

@pyoudeyer

6 years

Honoured to receive the prize @inria @AcadSciences from French ministry of research @VidalFrederique and get an opportunity to talk about how fundamental research on curiosity-driven learning and #AI can find societal applications in education #edtech

Inria

@Inria

6 years

#PrixInria2018 🏆 @VidalFrederique remet à @pyoudeyer le prix du jeune chercheur @Inria - @AcadSciences pour son activité de recherche autour de la curiosité artificielle

1

5

12

2

5

60

Pierre-Yves Oudeyer

@pyoudeyer

6 years

J. Gottlieb and I's @NatRevNeurosci review is out! "Towards a neuroscience of active sampling and curiosity" We cover converging work in #neuroscience , #psychology and #AI , giving rise to a new exciting field.

2

16

62

Pierre-Yves Oudeyer

@pyoudeyer

5 years

Lenia: Biology of Artificial Life Congrats @BertChakovsky for this publication in Complex Systems, outstanding work on designing and studying a continuous game of life ! Go play with it:

Lenia - Mathematical Life Forms

Lenia (from Latin lenis, “smooth”) is a cellular automaton, like Conway’s Game of Life, but with continuous states and continuous space-time. It supports a g...

www.youtube.com

1

21

57

Pierre-Yves Oudeyer

@pyoudeyer

6 years

Unsupervised RL workshop @iclr2019 Great to see many of the topics and techniques worked on by the developmental #AI and #robotics community for a decade finally come through within the #ML circles ! Comp. models of infant devlopment @iclr2020 ? 😄

1

17

59

Pierre-Yves Oudeyer

@pyoudeyer

6 years

Curiosity-driven learning machines and the emergence of communication Interview with @DAVID_LYNCH and myself on the Ergo-Robots #art #science project It integrates several strands of research in #artificialintelligence autonomous #machinelearning #AI

Curiosity-driven learning and the emergence of communication //...

The Ergo-Robot experiment is an art-science installation, where curiosity-driven learning robots spontaneously explore their environment, invent a new langua...

www.youtube.com

1

16

58

Pierre-Yves Oudeyer

@pyoudeyer

2 years

Instead of a #NeurIPS paper, what important societal challenge one could be working on?

12

2

57

Pierre-Yves Oudeyer

@pyoudeyer

1 year

📺Les vidéos de notre série pédagogique ( @FlowersINRIA @Inria ) expliquant les modèles de langage aux collégiens et lycéens sont disponibles en téléchargement 🚀🎉 #Litt éracie #IAG énérative ⬇️

Pierre-Yves Oudeyer

@pyoudeyer

2 years

Dans les collèges et lycées tout le monde parle de #ChatGPT 🦜🧑‍🎓🗣️ Mais combien savent comment ça fonctionne ?⚙️ Quelles sont ses possibilités et ses limites ? 🤔 Pour les enseignants et les élèves, voici un tour d'horizon en 5 mn 🚀🤖💡 #education

2

74

99

1

51

59

Pierre-Yves Oudeyer

@pyoudeyer

6 years

Not a bug !!! The robot discovered that tapping the joystick and scratching the table with hand would make table vibrations that move very slightly the ball ==> non zero reward ==> refining that behaviour !!! But a clear deceptive local minimum :) Here is the vibration !

1

6

57

Pierre-Yves Oudeyer

@pyoudeyer

4 years

Intrinsically Motivated Exploration of Learned Goal Spaces Congrats to A. Laversanne-Finot & A. Péré for the journal extension of this CorL paper studying how diversity search algorithms can work from pixel-level perception on clumsy real physical robots

Pierre-Yves Oudeyer

@pyoudeyer

6 years

Our new #Corl18 paper: How curiosity-driven autonomous goal setting enables to discover independantly controllable features of environments pdf: Blog: Colab: #machinelearning #AI #NeuralNetworks

6

130

305

1

19

57

Pierre-Yves Oudeyer

@pyoudeyer

2 years

How can an RL agent learn to use language to get help from a knowledge source ? How can it learn when and what to ask? Learn how to interpret the meaning of answers? 👉 See our new paper accepted at #ICML2022 📖 💻

4

8

56

Pierre-Yves Oudeyer

@pyoudeyer

7 years

Results of my #NIPS2017 poll on opinion of researchers about pace of scientific advances in #artificialintelligence : it seems many think we are not on the way to #intelligenceexplosion @fchollet :)

3

23

52

Pierre-Yves Oudeyer

@pyoudeyer

4 years

Several open #PhD & #postdoc positions in my team @FlowersINRIA : - emergent communication, curiosity-driven multi-agent Deep RL - adaptive training of attention for humans - Deep RL to optimize intervention in #Covid epidemic models #AI #MachineLearning

0

28

55

Pierre-Yves Oudeyer

@pyoudeyer

6 years

Y. Bengio calling #machinelearning to model infant development ! What a great path, looking back 10 years ago when #developmentalrobotics papers on intrinsic motivation and curriculum learning where thought to be crazy by the same #ml community #NeurIPS

4

21

53

Pierre-Yves Oudeyer

@pyoudeyer

6 years

Hey Twitter, I am searching for online educational material usable to introduce #AI and #machinelearning concepts to 12-15 years old in school. It can be full pedagogical activities, or (fun) interactive web simulation that afford educational explanations. All ideas welcome !

18

21

51

Pierre-Yves Oudeyer

@pyoudeyer

3 years

🌻Introducing TransFlower = #Transformer + Normalizing Flows for high-dim contin. control 👉 #Dance generation🕺💃 cond. on music wth various styles & motion prompting, congrats @guillefix ! Paper Code Fun

Transflower: Probabilistic autoregressive dance generation with...

Project page: https://metagen.ai/transflowerGuillermo Valle-Pérez, Gustav Eje Henter, Jonas Beskow, André Holzapfel, Pierre-Yves Oudeyer, Simon AlexandersonD...

www.youtube.com

1

13

54

Pierre-Yves Oudeyer

@pyoudeyer

1 year

Large language models are essentially cultural transmission technologies 🦜🗣️🦜🗣️🔁 They are culture models, encoding cultural regularities present in texts and in human feedback they were trained on. They amplify them when they are used, by humans or other machines #LLMs 🧵

1

19

54

Pierre-Yves Oudeyer

@pyoudeyer

4 years

Lenia and Expanded Universe #gameoflife in higher dimensions, featuring the emergence of "virtual eukaryotes" that reproduce, differentiate, migrate, interact and react with each other. Outstanding work by @BertChakovsky (independent researcher !)

Bert Chan

@BertChakovsky

4 years

New work on Lenia submitted to @ALifeConf 2020 is now on arXiv. Let me know what you think! (code will be updated soon) 1/4

2

27

107

0

15

52

Pierre-Yves Oudeyer

@pyoudeyer

6 years

Interested in autonomous goal setting and exploration? #ICLR2018 poster #24 4:30-6:30p today

1

14

49

Pierre-Yves Oudeyer

@pyoudeyer

3 years

Behavioral and Cognitive Robotics: An Adaptive Perspective An open-access book by Stefano Nolfi who has been a major actor of evolutionary and behavioral robotics

1

19

53

Pierre-Yves Oudeyer

@pyoudeyer

3 years

New blog post from @RPortelas : Teacher Algorithms for Deep RL Agents that Generalize in Procedurally Generated Environments

1

10

54

Pierre-Yves Oudeyer

@pyoudeyer

2 years

Autotelic AI: agents learn by sampling and pursuing their own goals Vygotskian AI: agents use language as a cognitive tool (inner speech) Vygotskian Autotelic AI: agents use language to imagine their own goals & for planning, reasoning and learning about them

2

4

52

Pierre-Yves Oudeyer

@pyoudeyer

4 years

PBCS: Efficient Exploration and Exploitation Using a Synergy between Reinforcement Learning and Motion Planning by Matheron, Perrin & @OlivierSigaud Extends Go-Explore, builds on Backplay and the idea of decoupling exploration and exploitation

1

14

49

Pierre-Yves Oudeyer

@pyoudeyer

6 years

Mind-blowing. What if the billions invested in #deeplearning were used to investigate morphogenesis? This was the route Alan Turing pointed out to understand the origins of intelligence.

Anatoly Buchin 💻🧠

@neurojoy

6 years

How does one cell become an organism? Check out this time lapse.

0

9

29

6

14

47

Pierre-Yves Oudeyer

@pyoudeyer

5 years

A key dimension of natural intelligence is in choosing what and when to learn (and whether the learning is done through curve fitting or something else is orthogonal). Beyond statistical learning, AI require advanced motivational/teleological architectures

Michael Nielsen

@michael_nielsen

5 years

Whenever I see this kind of headline, I always think "But what if intelligence is mostly about curve-fitting, and we're merely too un-self-aware to notice?"

26

69

447

3

11

50

Pierre-Yves Oudeyer

@pyoudeyer

6 years

I just found back my Epirob 2009 presentation discussing theoretical models of how curiosity and language learning are closely linked in children. Cool to see a posteriori how we implemented this vision with my team @FlowersINRIA in the last decade.

4

14

45

Pierre-Yves Oudeyer

@pyoudeyer

1 year

💡Curiosity-driven autotelic #AI can help scientists uncover behaviors of complex self-organized biological systems🔬 Here we show how they help reveal GRNs surprising (sensorimotor?) abilities👾🚀 With interactive paper, tutorials and open-source code:

Michael Levin

@drmichaellevin

1 year

New preprint! @mayalen_etc @Clement_MF_ @pyoudeyer - AI-driven Automated Discovery Tools Reveal Diverse Behavioral Competencies of Biological Networks. Next step in our efforts to study molecular networks with the lens of basal cognition.

10

39

198

2

15

51

Pierre-Yves Oudeyer

@pyoudeyer

6 years

The mysteries of curiosity-driven learning and extreme specialization in humans Great new dialog in the latest @ieeecis newsletter on Cognitive and Developmental Systems, initiated by @celestekidd and a great list of contributors @cogdevsoc @DevRobNews

3

13

47

Pierre-Yves Oudeyer

@pyoudeyer

4 years

Great that talks of #Neurips2020 Deep RL Workshop are online + accessible without registration, thanks organizers! Mine is about machines that invent their own problems, wth a focus on meta-diversity search and language as a tool for creative exploration

Deep RL Workshop, NeurIPS 2020

About In recent years, the use of deep neural networks as function approximators has enabled researchers to extend reinforcement learning techniques to solve increasingly complex control tasks. The...

sites.google.com

Pieter Abbeel

@pabbeel

4 years

Very excited about the @NeurIPSConf 2020 Deep RL Workshop! Amazing line-up of invited speakers: @marcgbellemare Matt Botvinick @RealAshEdwards Karen Liu @SusanMurphylab1 Anusha Nagabandi @pyoudeyer Peter Stone Paper submission deadline: Oct 5, 2020

2

60

282

1

14

48

Pierre-Yves Oudeyer

@pyoudeyer

6 years

Open #postdoc position @FlowersINRIA @inria Topic: Curiosity-driven deep reinforcement learning in open virtual worlds This project involves a collaboration with #Ubisoft Details: #machinelearning #AI #videogames @unity3d

2

27

48

Pierre-Yves Oudeyer

@pyoudeyer

3 years

So happy to give the inaugural talk of the "Developing Minds" online lectures, with Linda Smith and Josh Tenenbaum following soon! I will talk about Developmental #AI : machines that learn like children and help children learn better 👉 Register here (free attendance):

DevelopingMinds

@Dev_Minds

3 years

New "Developing Minds" global online lecture series launched. Inaugural lecture by @pyoudeyer on September 30th, 13:00 UTC. @Dev_Minds #DevelopingMinds #AI #DevelopmentalPsychology #Neuroscience #Robotics

2

10

24

1

9

47

Pierre-Yves Oudeyer

@pyoudeyer

2 years

So many people try to understand #LLMs from the lens of human-like intelligence (both people who over hype LLMs and those claiming they're superficial parrots 🦜- all comparing them to humans 🧑‍🤝‍🧑🧠) Maybe we should rather try to study them as alien intelligence ? 👾💡 🧵

7

14

49

Pierre-Yves Oudeyer

@pyoudeyer

4 years

A developmental robotics/computational development workshop @NeurIPSConf ! Amazing and so great !!! and with a great team of organizers ! "BabyMind: How babies learn and how machines can imitate" @Chen_Yu_CY any info available on deadlines?

1

13

47

Pierre-Yves Oudeyer

@pyoudeyer

6 years

Great introductory course on (Deep) #reinforcementlearning from @OlivierSigaud @ISIR_labo with Videos and slides: Jupyter lab notebooks: #neuralnetworks #machinelearning

Reinforcement Learning Class: Introduction

In this video, I introduce a whole sequence of several videos corresponding to a full RL class, then I explain what is reinforcement learning when compared t...

www.youtube.com

1

23

43

Pierre-Yves Oudeyer

@pyoudeyer

2 years

Large language models could be seen as interactive encyclopedias of human cultural habits (those which can be encoded in text) If trained on well-chosen corpuses, they could become a useful tool to address scientific questions about human cultures (provided adequate methodology)

2

4

48

Pierre-Yves Oudeyer

@pyoudeyer

4 years

On the topic of curriculum learning for #DeepRL , see also the great blog post of our co-author @lilianweng #AI #machinelearning #neuralnetworks

Curriculum For Reinforcement Learning

lilianweng.github.io

Pierre-Yves Oudeyer

@pyoudeyer

5 years

Automatic Curriculum Learning For Deep RL: A Short Survey What is ACL? Why use ACL? Which forms of ACL (e.g. goal generation, reward shaping, procedural gen. of envs, self-play)? What open questions? w @RPortelas @cedcolas @lilianweng @katjahofmann

2

50

157

1

20

46

Pierre-Yves Oudeyer

@pyoudeyer

4 years

Happy that our (short) review paper on Automatic Curriculum Learning in #DeepRL will be presented @IJCAIconf This field has an amazing dynamics! Updated version: #AI #neuralnetworks #deeplearning

Automatic Curriculum Learning For Deep RL: A Short Survey

Automatic Curriculum Learning (ACL) has become a cornerstone of recent successes in Deep Reinforcement Learning (DRL).These methods shape the learning trajectories of agents by challenging them...

arxiv.org

Pierre-Yves Oudeyer

@pyoudeyer

5 years

Automatic Curriculum Learning For Deep RL: A Short Survey What is ACL? Why use ACL? Which forms of ACL (e.g. goal generation, reward shaping, procedural gen. of envs, self-play)? What open questions? w @RPortelas @cedcolas @lilianweng @katjahofmann

2

50

157

1

11

47

Pierre-Yves Oudeyer

@pyoudeyer

11 months

Instead of asking whether #AI systems are (super)intelligent or (half)conscious, maybe we should ask whether they are useful, inspiring, enlightning, pedagogical, beautiful, poetic or simply fun

3

5

48

Pierre-Yves Oudeyer

@pyoudeyer

4 years

The IMAGINE approach to using language for creative exploration in #DeepRL agents will be presented next week @NeurIPSConf Camera ready: Videos: Code: Colab notebook: #neurips2020

IMAGINE.ipynb

Colaboratory notebook

colab.research.google.com

Tristan Karch

@karch_tristan

4 years

Excited to share our work: Language as a Cognitive Tool to Imagine Goals in Curiosity-Driven Exploration. Accepted at NeurIPS 2020! Link to paper:

4

46

236

0

4

42

Pierre-Yves Oudeyer

@pyoudeyer

1 year

A major challenge in current research in #LLMs 🦜 is how they can be grounded in interactive environments, aligning their pretrained knowledge to external dynamics 🤖🎮💡 ▶️Next week @icmlconf @ClementRomac & @CartaThomas2 present GLAM, an online RL approach to do this !!! 📖🚀

Clément ROMAC

@ClementRomac

2 years

🚨🚨 Large Language Models encode world knowledge that can benefit RL agents when solving decision-making problems (e.g. in robotics 🤖). But what happens when their knowledge are wrong? 🎯 Can we align LLMs and ground their knowledge through interactions with online RL? 🕹

3

44

230

1

11

43

Pierre-Yves Oudeyer

@pyoudeyer

5 years

Open #postdoc or #researchengineer position in my lab on curiosity-driven open-ended learning and exploration in large virtual worlds, in collaboration with @Ubisoft Application details: #machinelearning #AI #unity3d

researchEngineerPostdocUbisoft19.pdf

Shared with Dropbox

www.dropbox.com

3

22

43

Pierre-Yves Oudeyer

@pyoudeyer

1 year

Very clear and thoughtful paper on natural curiosity by philosopher Jennifer Nagel, with extended connections among works in philosophy, psychology and #AI

0

10

46

Pierre-Yves Oudeyer

@pyoudeyer

2 years

Thank you everyone for amazing contributions and discussions at this week's Dagstuhl seminar on #developmentalAI !!! This showed great new research opportunities and synergies at the crossroads of #AI #devpsychology #machinelearning 👶🤖💻🚀

1

17

46

Pierre-Yves Oudeyer

@pyoudeyer

4 years

"How does one measure the #intelligence of an #artificialintelligence ?" My answer here (also discussing curiosity in humans and machines, and #educational applications):

2

9

43

Pierre-Yves Oudeyer

@pyoudeyer

6 years

Interested by a PhD in deep #reinforcementlearning applied to adaptive language interaction for intelligent dialog agents? Application details here: #MachineLearning #neuralnetworks #deeplearning #NLP #language #web @CloudTemple

RCL-CT-THESE-interaction-manager-finalV2.pdf

Shared with Dropbox

www.dropbox.com

0

26

39

Pierre-Yves Oudeyer

@pyoudeyer

2 years

In this thread I'd like to share a very elegant idea proposed by Thomas Carta that he'll present this week @NeurIPSConf It's about automating task-specific reward shaping in language-conditioned RL without domain-specific expert knowledge 📖

3

12

43

Pierre-Yves Oudeyer

@pyoudeyer

3 years

To start the new year with fun, here's a gift from the @flowersInria team: 👉An interactive online app to test generalization capabilities of #DeepRL agents! 🎁🤖

Interactive web demo of generalization in Deep Reinforcement Learning

See https://developmentalsystems.org/Interactive_DeepRL_Demo/In this demo, all the available agents were trained using Soft Actor Critic as the DRL student a...

www.youtube.com

2

12

41

Pierre-Yves Oudeyer

@pyoudeyer

2 years

Congratulations ⁦ @Clement_MF_ ⁩ for the brilliant habilitation talk on the ecology of open-ended skill acquisition 👍🎉🚀

6

42

Pierre-Yves Oudeyer

@pyoudeyer

2 years

Chimpanzees 🐒, fishes 🐟, spiders 🕸️and bipedal walkers trained with DeepRL are now exploring the jungle in a @huggingface space 😀 👉 Here's the link to launch them in your browser and play ! Thanks a lot @ClementRomac !

Clément ROMAC

@ClementRomac

2 years

📢 We just integrated our Deep RL interactive demo in a @huggingface space 🤗👉 This version fetches policies uploaded on the hub with the tag 'teach-my-agent-parkour'. You can thus choose policies you want to play with and even upload yours! 1/9 🧵

1

12

84

0

8

40

Pierre-Yves Oudeyer

@pyoudeyer

2 years

A great perspective on the origins of play in humans, articulating various theories ranging from predictive processing to the learning progress hypothesis. It also includes a nice discussion of niche construction mechanisms in this context.

Marc Malmdorf Andersen

@Marc_M_Andersen

2 years

🥳🚨 PUBLICATION ALERT! 🚨 🥳 Have you ever wondered why humans play? How play relates to learning? Why play feels so fun? In this paper, we propose a cognitive theory of play utilising recent ideas of the brain as a predictive organ. OA: 🧵 1/5

18

155

582

1

6

40

Pierre-Yves Oudeyer

@pyoudeyer

11 months

🌐Building on our work on grounding #LLMs 🦜 in interactive environments with online RL, we are opening a master level 6 months internship @FlowersINRIA to study the impact of such grounding on the knowledge and structure of these LLMS 🚀 👉 To apply:

Clément ROMAC

@ClementRomac

2 years

🚨🚨 Large Language Models encode world knowledge that can benefit RL agents when solving decision-making problems (e.g. in robotics 🤖). But what happens when their knowledge are wrong? 🎯 Can we align LLMs and ground their knowledge through interactions with online RL? 🕹

3

44

230

1

7

41

Pierre-Yves Oudeyer

@pyoudeyer

5 years

Great work showing how intrinsically motivated goal exploration can be used to automate discovery of novel patterns in wet chemical systems !!! Congratulations @jgrizou and team! Intrinsically motivated #machinelearning algorithms open new perspectives in physics/chemistry

Jonathan Grizou

@jgrizou

5 years

We applied intrinsically motivated #AI to explore a chemical system in the lab: @ScienceAdvances Results: - 3x more varied observations than screening - enabled new observation that we study in depth - data show we improve rate of discoveries 1/8

1

11

34

2

27

41

Pierre-Yves Oudeyer

@pyoudeyer

3 years

Tomorrow at 5:05 pm GMT I'll be discussing with @joelbot3000 on open-endedness at the Never Ending RL workshop @iclr_conf #ICLR2021 That will be a lot of fun!

NERL2021

Motivation

sites.google.com

Never Ending RL

@neverendingrl

4 years

1/N We are excited to introduce our @iclr_conf workshop: A Roadmap to Never-Ending RL. We invite you to submit papers (up to 6 pages, excluding references and appendix) in the @iclr_conf format. Submission Deadline: February 26, 2021 #NERL2021 #ICLR2021

1

29

112

1

11

41

Pierre-Yves Oudeyer

@pyoudeyer

2 years

Large language models can be used as agent policies, and grounded using online RL to solve RL tasks in interactive environments 👉 How their prior knowledge increases sample efficiency? 👉 How can it boost generalization to new tasks? Study led by @ClementRomac @CartaThomas2 !

Clément ROMAC

@ClementRomac

2 years

🚨🚨 Large Language Models encode world knowledge that can benefit RL agents when solving decision-making problems (e.g. in robotics 🤖). But what happens when their knowledge are wrong? 🎯 Can we align LLMs and ground their knowledge through interactions with online RL? 🕹

3

44

230

0

8

40

Pierre-Yves Oudeyer

@pyoudeyer

3 years

Here's the talk I gave @EcoTheoryRL workshop @NeurIPSConf Sculpting human-like AI by sculpting social environments TL;DR 1) methods for educating learning machines are key and still mostly unexplored 2) we should put human-like sociality in AI envs 1/n

Sculpting human-like AI systems by sculpting their social environme...

Title: Sculpting human-like AI systems by sculpting their social environmentsTalk from Pierre-Yves Oudeyer for the workshop on Ecological Theory of RL Worksh...

www.youtube.com

2

7

39