top of page
Search
Randomize Everything. Just Don't Mix the Domains.
Domain randomization is supposed to stop a model from memorizing its surroundings. Our thermal-imaging experiment showed us exactly where that idea stops working. Synthetic data comes with a familiar rule: if you don't want the model to memorize something, randomize it. Change the lighting. Change the camera angle. Change the texture. Swap the background, again and again. Do it enough, and the model should stop caring about the scenery and start learning the object itself. Th
Finn Chen
Sep 224 min read
Not Just Relacing Real Data with Synthetic. Multiply It.
Synthetic data can sure replace real data in most cases, but adding a bit of real date can help calibrate and anchor the models for better results.
Finn Chen
Sep 45 min read
From Correlation to Causation: Why Pharma Is Embracing Causal AI
Why causal AI is becoming pharma’s next strategic differentiator — from target validation to clinical trials and real-world evidence. Visualizing the Ladder of Causation in Pharma: from seeing correlations, to doing interventions, to imagining counterfactuals — the foundation of causal AI in drug discovery. Introduction: The Limits of Correlation in Drug Discovery Drug discovery today generates unprecedented volumes of data. Omics, imaging, electronic health records, and real
Esther Chen
Sep 7, 20256 min read
bottom of page