of 1.2X • Outperforms SPX Index by 60X over period • Low turnover with average trade signal duration of 100 days • No in-sample optimization , Machine Learning techniques, Genetic Algos, or Feed Forward Neural Networks 70 - 60 a. • 50 0 • 40 5 a 30 20 10 BRKM —AAPL —SPX —Atorus - 1995 1996

l Ant Problem 224 33.3.5 Discussion 229 33.3.6 Conclusion 229 33.4 Integrating Feature Selection Into the Learning Process 230 33.4.1 Machine Learning, Feature Selection and AGI 231 33.4.2 Data- and Feature- Focusable Learning Problems 232 33.4.3 Integrating Feature Selection Into Learning 233

ique that does feature selection repeatedly during the MOSES learning process, rather than just doing it initially prior to MOSES learning. 33.4.1 Machine Learning, Feature Selection and AGI The relation between feature selection and machine learning appears an excellent example of the way that, even when the

lvin. The Cerebral Code. MIT Press, 1996. Car85. S. Carey. Conceptual Change in Childhood. MIT Press, 1985. Car97. R Caruana. Multitask learning. Machine Learning, 1997. Cas85. R. Case. Intellectual development: Birth to adulthood. Academic Press, 1985. Cas04. N. L. Cassimatis. Grammatical processing using

2006. William Calvin. The Cerebral Code. MIT Press, 1996. S. Carey. Conceptual Change in Childhood. MIT Press, 1985. R Caruana. Multitask learning. Machine Learning, 1997. R. Case. Intellectual development: Birth to adulthood. Academic Press, 1985. N. L. Cassimatis. Grammatical processing using the mechanisms o