Everardo González (egonzalez@geomar.de)
Naveen Parameswaran (nparameswaran@geomar.de)
\[N=R_{*}\cdot f_{\mathrm{p}}\cdot n_{\mathrm{e}}\cdot f_{\mathrm{l}}\cdot f_{\mathrm{i}}\cdot f_{\mathrm{c}}\cdot L\]
"But where is everybody?"
— Enrico Fermi, 1950
Chpt. 1
Uncertainty Quantification and Information Gain
One big question:
"Where to Sample Next?"
$D_{KL}(P\|Q) =$
$-\sum\limits_{x\in X}$
$p(x)$
$\log($
$q(x)$
$) + \sum\limits_{x\in X}$
$p(x)$
$\log($
$p(x)$
$)$
$D_{KL}(P\|Q) =$
$-\sum\limits_{x\in X}$
$p(x)$
$\log($
$q(x)$
$) + \sum\limits_{x\in X}$
$p(x)$
$\log($
$p(x)$
$)$
~400 Global Maps
~5000 Total Organic Carbon Measurements
Chpt. 2
Shap & Uncertainty Quantification
One day...
Previous Works: Uncertainty Quantification, SHAP, ...
InfoSHAP: Uncertainty Quantification x SHAP
🤔
Chpt. 3
Acoustic Embeddings for Underwater Ecosystem Monitoring
~50h recordings in German North Sea coastal waters
5 sec. snippets ⇒ ~40,000 labeled (!!!) datapoints
~20 unique feature combinations ⇒ classes
Solution: Word Embeddings!
Can we do that for underwater soundscapes?
contrastive learning with MobileNet 2
Chpt. 4
AI x Ocean Science: GEOMAR ML Seminar Series
Questions? Comments?
egonzalez@geomar.de