`

Timezone: »

 
Learning not to learn: Nature versus nurture in silico
Robert Lange

@ None

Animals are equipped with a rich innate repertoire of sensory, behavioral and motor skills, which allows them to interact with the world immediately after birth. At the same time, many behaviors are highly adaptive and can be tailored to specific environments by means of learning. In this work, we use mathematical analysis and the framework of meta-learning (or 'learning to learn') to answer when it is beneficial to learn such an adaptive strategy and when to hard-code a heuristic behavior. We find that the interplay of ecological uncertainty, task complexity and the agents' lifetime has crucial effects on the meta-learned amortized Bayesian inference performed by an agent. There exist two regimes: One in which meta-learning yields a learning algorithm that implements task-dependent information-integration and a second regime in which meta-learning imprints a heuristic or 'hard-coded' behavior. Further analysis reveals that non-adaptive behaviors are not only optimal for aspects of the environment that are stable across individuals, but also in situations where an adaptation to the environment would in fact be highly beneficial, but could not be done quickly enough to be exploited within the remaining lifetime. Hard-coded behaviors should hence not only be those that always work, but also those that are too complex to be learned within a reasonable time frame.

Author Information

Robert Lange (Einstein Center for Neurosciences)

I have finished my undergraduate studies in economics at the University of Cologne. During that time I have worked as a student research assistant for Prof. Alex Ludwig (Goethe University Frankfurt) and Prof. Helge Braun (University of Cologne). The projects mainly focused on public policy evaluation and the intersection of retirement and unemployment insurance systems. I developed a fascination for data wrangling and the computational aspects of Economics. Since September I am part of the 2017 cohort of the Data Science Master's Program at the Barcelona Graduate School of Economics. I am fully convinced that the intersection between behavioral sciences and statistical learning is crucial in order to improve almost every aspect of our lives. Therefore, I am looking forward to pursuing a second master's degree in the are of cognitive sciences and artificial intelligence coming next fall. At the moment my interest focuses on computational statistics and machine learning, interdisciplinary applications such as cognitive sciences, biometrics and the philosophy of risk.

More from the Same Authors

  • 2019 : Poster Presentations »
    Rahul Mehta · Andrew Lampinen · Binghong Chen · Sergio Pascual-Diaz · Jordi Grau-Moya · Aldo Faisal · Jonathan Tompson · Yiren Lu · Khimya Khetarpal · Martin Klissarov · Pierre-Luc Bacon · Doina Precup · Thanard Kurutach · Aviv Tamar · Pieter Abbeel · Jinke He · Maximilian Igl · Shimon Whiteson · Wendelin Boehmer · RaphaĆ«l Marinier · Olivier Pietquin · Karol Hausman · Sergey Levine · Chelsea Finn · Tianhe Yu · Lisa Lee · Benjamin Eysenbach · Emilio Parisotto · Eric Xing · Ruslan Salakhutdinov · Hongyu Ren · Anima Anandkumar · Deepak Pathak · Christopher Lu · Trevor Darrell · Alexei Efros · Phillip Isola · Feng Liu · Bo Han · Gang Niu · Masashi Sugiyama · Saurabh Kumar · Janith Petangoda · Johan Ferret · James McClelland · Kara Liu · Animesh Garg · Robert Lange