learning_rate=0.1,
n_estimators=2,
learning_rate=0.1,
n_jobs=-1,
colsample_bytree=0.05
)
# training the model
model.fit(X_train.iloc[:,2:], y_train.iloc[:,2:])
# saving model
model_pathname = Path(model_directory_path) / "model.joblib"
print(f"Saving model in {model_pathname}")
joblib.dump(model, model_pathname)
def infer(X_test: pd.DataFrame, model_directory_path: str = "resources") ->
pd.DataFrame:
# loading the model saved by the train function at previous iteration
model = joblib.load(Path(model_directory_path) / "model.joblib")
# creating the predicted label dataframe with correct dates and ids
y_test_predicted = X_test[["date", "id"]].copy()
y_test_predicted["value"] = model.predict(X_test.iloc[:, 2:])
return y_test_predicted
This groundbreaking Crunch has been designed by an
exceptional team of renowned experts: 2021 Nobel
Laureate in Economics Prof. Guido Imbens, Horst Simon,
former Deputy Director of the Berkeley Lab, and Prof.
Marcos Lopez de Prado, Global Head of Quantitative
R&D at ADIA.
Uncovering the causal
relationships between variables is a crucial step
towards making AI systems more transparent and truly
intelligent. By discovering the underlying causal
structure from observational data, Crunch helps its
customers to move beyond mere correlation and enable
machines to understand the 'why' behind
phenomena. This Crunch expects the community models to
unveil the causal directed acyclic graph (DAG) that
governs sets of variables, pushing the boundaries of
AI's reasoning capabilities. Such advancements are
vital across numerous fields, including healthcare,
economics, social sciences, and environmental studies.
X Alpha deploys capital in early and mid-stage
disruptive companies across the US and western
Europe.
For the first time ever, a leading
industry expert with a 15-year proven track record
collaborates with a Community of more than 5,000 data
scientists and 600 PhDs. This first of a kind
partnership aims to create a groundbreaking AI-driven
prototype for venture capital.
VC firms often
make investment decisions based on insufficient
information and heavily depend on human intuition and
biased decision-making.
The CrunchAI
Machine Learning Community will identify trends,
relationships and hidden patterns leading to replicable,
reproducible and unbiased Alpha-generating process for
Venture Capital.
DataCrunch leverages CrunchAI's quantitative
research to manage its systematic market-neutral
portfolio. The proprietary dataset encompasses thousands
of publicly traded U.S. companies, providing a
comprehensive view of the market landscape.
To
achieve this, DataCrunch requires the community to build
algorithms that can predict the relative performance of
assets within an investment universe. Specifically, the
community models should precisely rank the constituents
of this very universe.
Successful models
are expected to demonstrate consistent performance,
out-of-sample, offering valuable insights that can be
integrated into any systematic investment process.
"I'm very excited to see what the participants are going to come up with, because if they come up with useful things, that's going to be very impactful"
"Crowdsourcing has a very important role to play in investing. Firms turn investing problems into forecasting problems, then outsource to global researchers"
"Institutional finance hasn’t yet had disruption, but likely will; specifically with respect to the competition for research talent in the years to come"
