Alfredo Canziani @alfcnz.bsky.social · 02/07/2025To compute the movement of the state x(t), we need to temporally integrate its velocity field ẋ(t). 🤓 The control signal steering angle stays at 0, then 0.05π, then linearly to −0.20π. The vehicle moves along circumferences. Finally, a sweep of initial velocity is performed. 0100
Alfredo Canziani @alfcnz.bsky.social · 30/06/2025Currently, writing chapter 10, «Planning and control». Physical constrains for the evolution of the state (e.g. pure rotation of the wheels) are encoded through the velocity of the state ẋ = dx(t)/dt, a function of the state x(t) and the control u(t). 280
Alfredo Canziani @alfcnz.bsky.social · 09/06/2025Releasing the Energy-Book 🔋 from its first appendix's chapter, where I explain how I create my figures. 🎨 Feel free to report errors via the issues' tracker, contribute to the exercises, and show me what you can draw, via the discussion section. 🥳 github.com/Atcold/Energ... 0243
Alfredo Canziani @alfcnz.bsky.social · 07/06/2025On a summer Friday night, the first chapter sees the light. 🥹🥹🥹 170
Alfredo Canziani @alfcnz.bsky.social · 29/05/2025Yeah, it took me 20 days to get back 🥹🥹🥹 I swear I respond to instant messages as they get through! 🥲🥲🥲 Anyhow, one more successful semester completed. 🥳🥳🥳 080
Alfredo Canziani @alfcnz.bsky.social · 09/04/2025In this lecture from my new undergrad course, we review linear multiclass classification, leverage backprop and gradient descent to learn a linearly separable feature vector for the input, and observe the training dynamics in a 2D embedding space. 🤓 youtu.be/saskQ-EjCLQ 1120
Alfredo Canziani @alfcnz.bsky.social · 08/04/2025Training of a 2 → 100 → 2 → 5 fully connected ReLU neural net via cross-entropy minimisation. • it starts outputting small embeddings • around epoch 300 learns an identity function • takes 1700 epochs more to unwind the data manifold 1151
Reposted by Alfredo Canzianikhipu.ai @khipu-ai.bsky.social · 11/03/2025Did you enjoy Alfredo Canziani's lecture as much as we did?! If so, check out his website to find more about his educational offer: atcold.github.io You can also find really cool material on Alfredo's YouTube channel! @alfcnz.bsky.socialatcold.github.ioAlfredo Canziani 042
Reposted by Alfredo Canzianikhipu.ai @khipu-ai.bsky.social · 07/03/2025📣 A pocos días del comienzo del Khipu 2025, nos complace anunciar que tanto las actividades del salón principal como el acto de clausura del viernes se retransmitirán en directo por este canal: khipu.ai/live/. ¡Los esperamos!khipu.ai 062
Alfredo Canziani @alfcnz.bsky.social · 11/03/2025I *really* had a blast giving this improvised lecture on a topic requested on the spot without any sleep! 🤪 The audience seemed enjoying the show. 😄 To find more about my educational offer, check out my website! atcold.github.io Follow here and subscribe on YouTube! 😀 031
Alfredo Canziani @alfcnz.bsky.social · 11/02/2025In today's episode, we review the concepts of loss ℒ(𝘄, 𝒟), per-sample loss L(𝘄, x, y), binary cross-entropy cost ℍ(y, ỹ) = y softplus(−s) + (1−y) softplus(s), ỹ = σ(𝘄ᵀ𝗳(x)). Then, we minimised the loss by choosing convenient values for our weight vector 𝘄. @nyucourant.bsky.social 0172
Alfredo Canziani @alfcnz.bsky.social · 30/01/2025Tue morning: *prepares slides* Tue class: *improv blackboard lecture* Outcome: unexpectedly great lecture. Thu morning: *prep handwritten notes* Thu class: *executes blackboard lecture* Students: 🤩🤩🤩🤩🤩🤩🤩🤩🤩 @nyucourant.bsky.social @nyudatascience.bsky.social 2181
Alfredo Canziani @alfcnz.bsky.social · 23/01/2025I think the new undergrad course is going well. At least we're having fun! 😁😁😁youtu.beCosine (extract from undergrad DLSP25)YouTube video by Alfredo Canziani (冷在) 1161
Reposted by Alfredo CanzianiConnor Lindsey @connorlindsey.dev · 20/12/2024If you're interested in creative coding or simulations, come read The Nature of Code with us! I'm setting up a chill book club to read the book and share projects starting in January. DM me for info! 4668
Reposted by Alfredo CanzianiHector Padilla @elhector.bsky.social · 10/12/2024I was reading about Conway's Game of Life, and this whole amazing book is available online! The Nature of Code - by Daniel Shiffman Thank you @shiffman.net for making it free for the web! People, please consider buying a copy of the book :) natureofcode.com/cellular-aut...natureofcode.com7. Cellular AutomataIn Chapter 5, I defined a complex system as a network of elements with short-range relationships, operating in parallel, that exhibit emergent behavio 0355
Alfredo Canziani @alfcnz.bsky.social · 09/01/2025It takes me weeks to detox from a stressful work mindset and finally sleep in — without waking at dawn, driven by the haste of unfinished tasks. I struggle to turn on my laptop to do any work. When I do restart, the anxiety comes back instantaneously. Is this common? ☹️☹️☹️ 4200
Alfredo Canziani @alfcnz.bsky.social · 07/01/2025Preparing an “Intro to Deep Learning” blackboard undergraduate course. Number of currently enrolled students: zero. 🥹🥹🥹 Current motivation: very weak. 3120
Reposted by Alfredo CanzianiMan in Business Suit Levitating @dfeldman.org · 25/12/2024You’re still arguing about tabs vs. spaces? May I present… 15652621266
Reposted by Alfredo CanzianiKeenan Crane @keenancrane.bsky.social · 09/12/2024Entropy is one of those formulas that many of us learn, swallow whole, and even use regularly without really understanding. (E.g., where does that “log” come from? Are there other possible formulas?) Yet there's an intuitive & almost inevitable way to arrive at this expression. 22543128
Alfredo Canziani @alfcnz.bsky.social · 16/12/2024Vancouver’s cannellés are delicious! 😋 Get one at Granville Island public market! 3140
Alfredo Canziani @alfcnz.bsky.social · 14/12/2024UltraPixel: Advancing Ultra-High-Resolution Image Synthesis to New Peaks 030
Alfredo Canziani @alfcnz.bsky.social · 14/12/2024Scalable optimisation in the modular norm. Direct to the point. 😀😀😀 160
Alfredo Canziani @alfcnz.bsky.social · 14/12/2024Scaling continuous latent variable models as probabilistic integral circuits @nolovedeeplearning.bsky.social @gengala 1143
Reposted by Alfredo CanzianiHeiga Zen (全 炳河) @heigazen.bsky.social · 13/12/20241/2: ✨ Google DeepMind Tokyo is growing! ✨ We're building the future of AI with Project Astra. Join my team! We're hiring a Software Engineering Manager to lead the challenge from Japan 🇯🇵 boards.greenhouse.io/deepmind/job... #GoogleDeepMind #AI #Jobs #Tokyoboards.greenhouse.ioSoftware Engineering Manager, Multimodal Assistive Agents (Astra) - JapanTokyo, Japan 174
Alfredo Canziani @alfcnz.bsky.social · 13/12/2024Me: what’s your poster about? Presenter (P): let me tell you about the theory of constrained optimisation of convex functions. Me: no, what was the goal? P: okay, so the framework of blah and blah has the KKT blah and blah blah blah conditions and… Me: high level, plz. 2100
Alfredo Canziani @alfcnz.bsky.social · 13/12/2024Optimization Algorithm Design via Electric Circuits @StephenPBoyd1 1141
Alfredo Canziani @alfcnz.bsky.social · 13/12/2024Convolutional Differentiable Logic Gate Networks @FHKPetersen 3453
Alfredo Canziani @alfcnz.bsky.social · 13/12/2024An Eye for an Ear: Zero-shot Audio Description Leveraging an Image Captioner using Audiovisual Distribution Alignment 061
Alfredo Canziani @alfcnz.bsky.social · 13/12/2024Outstanding presentation, finally! DynaMo: In-Domain Dynamics Pretraining for Visuo-Motor Control @jeffacce.bsky.social @lerrelpinto.com 0112
Alfredo Canziani @alfcnz.bsky.social · 13/12/2024Meta 3D AssetGen: Text-to-Mesh Generation with High-Quality Geometry, Texture, and PBR Materials 070
Alfredo Canziani @alfcnz.bsky.social · 13/12/2024BetterBench: Assessing Al Benchmarks, Uncovering Issues, and Establishing Best Practices @ankareuel.bsky.social 190
Alfredo Canziani @alfcnz.bsky.social · 13/12/2024Hey, @neuripsconf.bsky.social, posters on the penultimate panel side are invisible. 🫥 There’s no light! This is extremely unfair for the presenter! See brightness comparison below. 👇🏻 161
Alfredo Canziani @alfcnz.bsky.social · 13/12/2024Met the twins @roman-bushuiev.bsky.social @anton-bushuiev.bsky.social 0142
Reposted by Alfredo CanzianiFelix Petersen @petersen.ai · 17/11/2024If you're excited about #AI with #logic, check out our fully animated video on YouTube: youtu.be/FKQfMwFZvIEyoutu.beConvolutional Differentiable Logic Gate Networks - NeurIPS Oral - difflogicYouTube video by Felix Petersen 0203
Reposted by Alfredo CanzianiFelix Petersen @petersen.ai · 04/12/2024Have you ever wondered how training dynamics differ between LLMs 🖋️ and Vision 👁️ models? We explore this and close the gap between VMs and LLMs in our #NeurIPS2024 paper "TrAct: Making First-layer Pre-Activations Trainable". Paper link 📜: arxiv.org/abs/2410.23970 Video link 🎥: youtu.be/ZjTAjjxbkRY 🧵 192
Reposted by Alfredo CanzianiFelix Petersen @petersen.ai · 17/11/2024Excited to share our #NeurIPS 2024 Oral, Convolutional Differentiable Logic Gate Networks, leading to a range of inference efficiency records, including inference in only 4 nanoseconds 🏎️. We reduce model sizes by factors of 29x-61x over the SOTA. Paper: arxiv.org/abs/2411.04732 29918
Alfredo Canziani @alfcnz.bsky.social · 12/12/2024Neural Assets: 3D-Aware Multi-Object Scene Synthesis with Image Diffusion Models @tkipf.bsky.social 020
Alfredo Canziani @alfcnz.bsky.social · 12/12/2024TrAct: Making First-layer Pre-Activations Trainable @FHKPetersen 070
Alfredo Canziani @alfcnz.bsky.social · 12/12/2024Lorentz-Equivariant Geometric Algebra Transformers for High-Energy Physics 061
Alfredo Canziani @alfcnz.bsky.social · 12/12/2024MoE Jetpack: From Dense Checkpoints to Adaptive Mixture of Experts for Vision Tasks 090
Alfredo Canziani @alfcnz.bsky.social · 12/12/2024The Practitioner's Guide to Continual Multimodal Pretraining @dziadzio.bsky.social @confusezius.bsky.social @vishaalurao.bsky.social @bayesiankitten.bsky.social 0244
Alfredo Canziani @alfcnz.bsky.social · 12/12/2024Newton Losses: Using Curvature Information for Learning with Differentiable Algorithms @FHKPetersen 1103