e/acc

How much does it cost to back up solar with batteries?

This turns out to be a tricky question. Here's my current, partial understanding

Links and short notes, 2025-03-18

The 150-year history of d/acc, a $120M progress fund, fertility on demand, 19th-century kids' sci-fi, the real lesson of Cassandra’s curse, giving the Devil the benefit of law, and more

Adversarial Data Collection: Human-Collaborative Perturbations for Efficient and Robust Robotic Imitation Learning

The pursuit of data efficiency, where quality outweighs quantity, has emerged as a cornerstone in robotic manipulation, especially given the high costs associated with real-world data collection. We propose that maximizing the informational density of individual demonstrations can dramatically reduce reliance on large-scale datasets while improving task performance. To this end, we introduce Adversarial Data Collection, a Human-in-the-Loop (HiL) framework that redefines robotic data acquisition through real-time, bidirectional human-environment interactions. Unlike conventional pipelines that passively record static demonstrations, ADC adopts a collaborative perturbation paradigm: during a single episode, an adversarial operator dynamically alters object states, environmental conditions, and linguistic commands, while the tele-operator adaptively adjusts actions to overcome these evolving challenges. This process compresses diverse failure-recovery behaviors, compositional task variations, and environmental perturbations into minimal demonstrations. Our experiments demonstrate that ADC-trained models achieve superior compositional generalization to unseen task instructions, enhanced robustness to perceptual perturbations, and emergent error recovery capabilities. Strikingly, models trained with merely 20% of the demonstration volume collected through ADC significantly outperform traditional approaches using full datasets. These advances bridge the gap between data-centric learning paradigms and practical robotic deployment, demonstrating that strategic data acquisition, not merely post-hoc processing, is critical for scalable, real-world robot learning. Additionally, we are curating a large-scale ADC-Robotics dataset comprising real-world manipulation tasks with adversarial perturbations. This benchmark will be open-sourced to facilitate advancements in robotic imitation learning.

- YouTube

Enjoy the videos and music you love, upload original content, and share it all with friends, family, and the world on YouTube.

2025 EthSF '25 Week d/acc Day

2025 EthSF '25 Week d/acc DayThursday, Mar 13, 2025Co-hosted by Berkeley RDI

<d/acc day> @ UC Berkeley · Luma

<d/acc day> 🗓 13 March, 2025 📍 UC Berkeley Led by Vitalik Buterin, d/acc is his vision for a more thoughtful approach to technological acceleration—one that…

Links and short notes, 2025-03-10

A visible sonic boom, AI in SaaS, progress on permitting reform, fewer tornado deaths, the closing of the frontier, Rousseau and Kant vs. the Age of Reason, and more

Solutionism, part 2

Chapter 5 of The Techno-Humanist Manifesto

Links and short notes, 2025-03-03

Progress studies in university and high school, a positive supply shock for truth, the hinge of history, Dean Ball on AI liability, Kevin Kelly on the handoff to bots, and more

Progress Conference 2025

A four-day event to connect people & ideas in the progress movement. October 16-19, 2025 in Berkeley, California

Towards an AI co-scientist

Scientific discovery relies on scientists generating novel hypotheses that undergo rigorous experimental validation. To augment this process, we introduce an AI co-scientist, a multi-agent system built on Gemini 2.0. The AI co-scientist is intended to help uncover new, original knowledge and to formulate demonstrably novel research hypotheses and proposals, building upon prior evidence and aligned to scientist-provided research objectives and guidance. The system's design incorporates a generate, debate, and evolve approach to hypothesis generation, inspired by the scientific method and accelerated by scaling test-time compute. Key contributions include: (1) a multi-agent architecture with an asynchronous task execution framework for flexible compute scaling; (2) a tournament evolution process for self-improving hypotheses generation. Automated evaluations show continued benefits of test-time compute, improving hypothesis quality. While general purpose, we focus development and validation in three biomedical areas: drug repurposing, novel target discovery, and explaining mechanisms of bacterial evolution and anti-microbial resistance. For drug repurposing, the system proposes candidates with promising validation findings, including candidates for acute myeloid leukemia that show tumor inhibition in vitro at clinically applicable concentrations. For novel target discovery, the AI co-scientist proposed new epigenetic targets for liver fibrosis, validated by anti-fibrotic activity and liver cell regeneration in human hepatic organoids. Finally, the AI co-scientist recapitulated unpublished experimental results via a parallel in silico discovery of a novel gene transfer mechanism in bacterial evolution. These results, detailed in separate, co-timed reports, demonstrate the potential to augment biomedical and scientific discovery and usher an era of AI empowered scientists.

- YouTube

Enjoy the videos and music you love, upload original content, and share it all with friends, family, and the world on YouTube.

Links and short notes, 2025-02-17

I go away for three weeks and everyone announces everything

First Annual Progress Conference - 2024

Progress Conference 2024 brought together top thinkers and doers in the progress movement. This playlist showcases our keynote speakers at the event. Watch t...

x.com

Solutionism, part 1

Chapter 5 of The Techno-Humanist Manifesto

x.com

x.com

The Future of Humanity is in Management

AI can dramatically expand human agency