A Toy Model of Universality: Reverse Engineering How Networks Learn Group Operations

Deep neural networks come in many sizes and architectures. The choice of architecture, in conjunction with the dataset and learning algorithm, is commonly understood to affect the learned neural representations. Yet, recent results have shown that different architectures learn representations with striking qualitative similarities. Here we derive an effective theory of representation learning under the assumption that the encoding map from input to hidden representation and t...

Though modern neural networks have achieved impressive performance in both vision and language tasks, we know little about the functions that they implement. One possibility is that neural networks implicitly break down complex tasks into subroutines, implement modular solutions to these subroutines, and compose them into an overall solution to a task - a property we term structural compositionality. Another possibility is that they may simply learn to match new inputs to lea...

Do neural networks, trained on well-understood algorithmic tasks, reliably rediscover known algorithms for solving those tasks? Several recent studies, on tasks ranging from group arithmetic to in-context linear regression, have suggested that the answer is yes. Using modular addition as a prototypical problem, we show that algorithm discovery in neural networks is sometimes more complex. Small changes to model hyperparameters and initializations can induce the discovery of q...

Neural network models have achieved high performance on a wide variety of complex tasks, but the algorithms that they implement are notoriously difficult to interpret. In order to understand these algorithms, it is often necessary to hypothesize intermediate variables involved in the network's computation. For example, does a language model depend on particular syntactic properties when generating a sentence? However, existing analysis tools make it difficult to test hypothes...

We introduce the HyperCube network, a novel approach for autonomously discovering symmetry group structures within data. The key innovation is a unique factorization architecture coupled with a novel regularizer that instills a powerful inductive bias towards learning orthogonal representations. This leverages a fundamental theorem of representation theory that all compact/finite groups can be represented by orthogonal matrices. HyperCube efficiently learns general group oper...

We consider the problem of discovering subgroup $H$ of permutation group $S_{n}$. Unlike the traditional $H$-invariant networks wherein $H$ is assumed to be known, we present a method to discover the underlying subgroup, given that it satisfies certain conditions. Our results show that one could discover any subgroup of type $S_{k} (k \leq n)$ by learning an $S_{n}$-invariant function and a linear transformation. We also prove similar results for cyclic and dihedral subgroups...

There is a concerted effort to build domain-general artificial intelligence in the form of universal neural network models with sufficient computational flexibility to solve a wide variety of cognitive tasks but without requiring fine-tuning on individual problem spaces and domains. To do this, models need appropriate priors and inductive biases, such that trained models can generalise to out-of-distribution examples and new problem sets. Here we provide an overview of the ha...

The rise of Artificial Intelligence (AI) recently empowered researchers to investigate hard mathematical problems which eluded traditional approaches for decades. Yet, the use of AI in Universal Algebra (UA) -- one of the fields laying the foundations of modern mathematics -- is still completely unexplored. This work proposes the first use of AI to investigate UA's conjectures with an equivalent equational and topological characterization. While topological representations wo...

A common assumption about neural networks is that they can learn an appropriate internal representations on their own, see e.g. end-to-end learning. In this work we challenge this assumption. We consider two simple tasks and show that the state-of-the-art training algorithm fails, although the model itself is able to represent an appropriate solution. We will demonstrate that encouraging an appropriate internal representation allows the same model to solve these tasks. While ...

Neural networks often exhibit emergent behavior, where qualitatively new capabilities arise from scaling up the amount of parameters, training data, or training steps. One approach to understanding emergence is to find continuous \textit{progress measures} that underlie the seemingly discontinuous qualitative changes. We argue that progress measures can be found via mechanistic interpretability: reverse-engineering learned behaviors into their individual components. As a case...