Posts

If you’re here from the QR code — welcome. The poster gave you the summary. This page shows the method working. The core intuition: jolts The poster mentions HiPPO representations, but doesn’t have room to explain why they’re useful for anomaly detection. Here’s the idea. HiPPO continuously compresses a stream of observations into a compact coefficient vector — essentially a polynomial approximation of the recent history of the signal. When something unexpected happens — a sudden offset, a spike, a slow drift — the update to that coefficient vector is abruptly large. We call these jolts. ...

The following content has been created as part of a Lecture series @HPI Potsdam. Download the notebook: Notebook.zip Show Document in new tab: /slides/information-retrieval-1

uflt-showcase.mp4 Your browser does not support the video tag. UFLT (Ultra Fast Labeling Tool) Label, enhance, interpret embeddable data quickly. Why this could be interesting for you When working with unlabeled data we as researchers often times have to decide to either: spend the time tediously labeling, or spend the money / resources to get our data labeled. In my current project I had a bunch of images I needed to label – in this case classify – quickly. Not all data needed to be labeled perfectly from the get-go. So I wanted to label as I go, deciding what is worth the effort along the way. I found the tooling available to me lacking and cumbersome, so I built a tool for myself: UFLT. My guess is that others like you could also profit from this tool. ...

The following content has been created as part of a Lecture series @HPI Potsdam. It is copied verbatim from the respective notebook. The Notebook Show Document in new tab: math4ml-gradient-slides.html

One two three. This just tests some stuff. Lists Interesting Math: $$ \frac{1}{3} $$Code: def hello(): print("World") External Code: # /// script # dependencies = [ # "rootpath", # ] # /// import rootpath rootpath.append() print("Hello Static") External Code Relative: # /// script # dependencies = [ # "rootpath" # ] # /// import rootpath rootpath.append() print("Hello Relative")