Plant-Inspired AI: Plants as Inspiration for Novel Problem Formulations, and Two Case Studies
Authors: Deepayan Sanyal, Joel Michelson, Carla E. Cao, Adam B. Roddy, Maithilee Kunda
Organizations: Department of Environmental Studies, New York University, New York City, NY, 10003, USA · Department of Philosophy, Universidad de Murcia, Murcia, 30010, Spain · School of Informatics, University of Edinburgh, Edinburgh, Scotland, EH8 9AB, UK
Artificial Intelligence (AI) has long been inspired by studies of biological intelligence. Reinforcement learning, for instance, drew inspiration from studies involving animal learning and is now a powerful paradigm for solving many real-world problems. Recently, plant biologists have uncovered a wide range of complex behaviors in plants that enable them to flexibly adapt to variable environments. Here, we argue that such behavior can motivate new AI frameworks encompassing a range of problems overlooked by existing problem-solving frameworks such as supervised learning, tree search, and constraint satisfaction. We illustrate this idea with two examples of intelligent problem-solving in plants: (1) leaf mimicry in Boquila trifoliolata, a vine capable of altering its leaves' morphology to resemble those of multiple host trees simultaneously; and (2) coordinated root-shoot growth, wherein plants allocate resources across organ systems exploring distinct environments. While leaf mimicry is highly specific to Boquila, coordination of root-shoot growth is shared across most plants. For both examples, we capture underlying computational principles and identify problems fitting these frameworks that are currently unaddressed by AI. Finally, we outline preliminary task formulations and discuss how these formulations may be applied to non-plant problems.
Figures & tables
Figure 1: Overview of the two case studies. (a) Examples of leaf mimicry in B. trifoliolata ( Gianoli and Carrasco-Urra, 2014 ) . Host tree leaves are marked with T , while the Boquila leaves are marked with V . (b) The ratio of auxin to cytokinin determines the extent to which roots and shoots grow in the plant ( Kurepa et al., 2023 ) . Different environmental conditions result in different optimal ratios.
Figure 2: Instantiation of the collective mimicry task described in Section 3.3 . The left image is a target to be imitated, sampled from the MNIST dataset ( LeCun et al., 1998 ) . The second image shows a cell graph G that might have been grown in an attempt to solve this task instance. This graph is not shaped like the digit depicted in T , so it is not a particularly good solution in this case. Edges between cells indicate the neighbors whose hidden states each cell may observe, while dotted arrows show where new cells may be grown. The central image shows the receptive fields of the cells; these fields depend on each cell’s position relative to contiguous cells in its row and column. The fourth image shows the mean value of each field; these are the cells’ observations of the target. Ultimately, the goal of the cells is to grow into the shape of the target image; the rightmost image shows the error in the appearance of the grown cell structure compared with the target T .
Figure 3: Overview of the root-shoot collaboration task. (a) The environment is divided into aboveground and belowground regions with different nutrient distributions. A shoot cell S can acquire Carbon and Sunlight, while a root cell R can acquire Nitrogen and Water. (b) Cells sense and store locally available nutrients, transfer nutrients to neighboring cells, and duplicate when growth requirements are met. (c) These local actions produce coordinated growth of the root and shoot over time, with growth being measured by the plant size N and balance B .
Artificial Intelligence (AI) algorithms frequently learn creative and unexpected solutions, surprising even expert researchers who develop and study them. They often astonish practitioners by discovering unanticipated behavior, exploiting loopholes in reward signals, or spontaneously uncovering previously unknown scientific phenomena. However, accounts of such unconventional behavior across machine learning are seldom formally documented. This work presents 26 curated firsthand anecdotes from various machine learning subfields representing the work of over 100 researchers. These anecdotes showcase the capability of modern AI systems to circumvent human-imposed design limitations and discover unexpected solutions to the tasks we train them on. Furthermore, these accounts are particularly important for the safety of future AI systems. They illustrate the fundamental challenge of aligning models with human values without diminishing their creativity, so they can make surprising discoveries without producing surprising, potentially harmful outcomes. The paper first details AI achieving superhuman success through reinforcement learning across many challenging domains. However, reward-driven optimization can fail when the model learns to hack an underspecified reward or unarticulated constraint. We then present case studies suggesting that harnessing internet-scale foundation models (FMs) has not resolved these fundamental challenges and, in fact, can supercharge them. Nevertheless, we argue that these same learning dynamics can be harnessed to accelerate scientific discovery. Finally, we hope this work provides a consolidated resource to inform future research and demonstrates that the tendency toward unexpected behaviors is commonplace in modern AI, highlighting the need to anticipate and manage AI's capacity for innovative, yet unpredictable, solutions. (abstract abridged)
Aaron Dharna, Cong Lu, Ryan Sullivan +3
University of British Columbia · Vector Institute · Recursive +3
We critically examine the limitations of current AI models in achieving autonomous learning and propose a learning architecture inspired by human and animal cognition. The proposed framework integrates learning from observation (System A) and learning from active behavior (System B) while flexibly switching between these learning modes as a function of internally generated meta-control signals (System M). We discuss how this could be built by taking inspiration on how organisms adapt to real-world, dynamic environments across evolutionary and developmental timescales.
Emmanuel Dupoux, Yann LeCun, Jitendra Malik
FAIR at META · École des Hautes Études en Sciences Sociales · NYU +1
The remarkable performance of modern AI systems has been driven by unprecedented scales of data, computation, and energy -- far exceeding the resources required by human intelligence. This disparity highlights the need for new guiding principles and motivates drawing inspiration from the fundamental organizational principles of brain computation. Among these principles, modularity has been shown to be critical for supporting the efficient learning and strong generalization abilities consistently exhibited by humans. Furthermore, modularity aligns well with the No Free Lunch Theorem, which highlights the need for problem-specific inductive biases and motivates architectures composed of specialized components that solve subproblems. However, despite its fundamental role in natural intelligence and its demonstrated benefits across a range of seemingly disparate AI subfields, modularity remains relatively underappreciated in mainstream AI research. In this work, we review several research threads in artificial intelligence and neuroscience through a conceptual framework that highlights the central role of modularity in supporting both artificial and natural intelligence. In particular, we examine what computational advantages modularity provides, how it has emerged as a solution across several AI research areas, which modularity principles the brain exploits, and how modularity can help bridge the gap between natural and artificial intelligence.
Alessandro Salatiello
Department of Computer Science, University of Tübingen