arXiv subjects

Mateja Jamnik

Publications and source records attributed to Mateja Jamnik.

77 records · Page 5Linked to original sources

arXiv2020

Towards Graph Representation Learning in Emergent Communication

Recent findings in neuroscience suggest that the human brain represents information in a geometric structure (for instance, through conceptual spaces). In order to communicate, we flatten the complex representation of entities and their attributes into a single word or a sentence. In this paper we use graph convolutional networks to support the evolution of language and cooperation in multi-agent systems. Motivated by an image-based referential game, we propose a graph referential game with varying degrees of complexity, and we provide strong baseline models that exhibit desirable properties in terms of language emergence and cooperation. We show that the emerged communication protocol is robust, that the agents uncover the true factors of variation in the game, and that they learn to generalize beyond the samples encountered during training.

cs.LG↗

arXiv2019

Decoupling feature propagation from the design of graph auto-encoders

We present two instances, L-GAE and L-VGAE, of the variational graph auto-encoding family (VGAE) based on separating feature propagation operations from graph convolution layers typically found in graph learning methods to a single linear matrix computation made prior to input in standard auto-encoder architectures. This decoupling enables the independent and fixed design of the auto-encoder without requiring additional GCN layers for every desired increase in the size of a node's local receptive field. Fixing the auto-encoder enables a fairer assessment on the size of a nodes receptive field in building representations. Furthermore a by-product of fixing the auto-encoder design often results in substantially smaller networks than their VGAE counterparts especially as we increase the number of feature propagations. A comparative downstream evaluation on link prediction tasks show comparable state of the art performance to similar VGAE arrangements despite considerable simplification. We also show the simple application of our methodology to more challenging representation learning scenarios such as spatio-temporal graph representation learning.

cs.LG↗

arXiv2019

Bayesian Optimisation with Gaussian Processes for Premise Selection

Heuristics in theorem provers are often parameterised. Modern theorem provers such as Vampire utilise a wide array of heuristics to control the search space explosion, thereby requiring optimisation of a large set of parameters. An exhaustive search in this multi-dimensional parameter space is intractable in most cases, yet the performance of the provers is highly dependent on the parameter assignment. In this work, we introduce a principled probablistic framework for heuristics optimisation in theorem provers. We present results using a heuristic for premise selection and The Archive of Formal Proofs (AFP) as a case study.

cs.AI↗

arXiv2019

Unsupervised and interpretable scene discovery with Discrete-Attend-Infer-Repeat

In this work we present Discrete Attend Infer Repeat (Discrete-AIR), a Recurrent Auto-Encoder with structured latent distributions containing discrete categorical distributions, continuous attribute distributions, and factorised spatial attention. While inspired by the original AIR model andretaining AIR model's capability in identifying objects in an image, Discrete-AIR provides direct interpretability of the latent codes. We show that for Multi-MNIST and a multiple-objects version of dSprites dataset, the Discrete-AIR model needs just one categorical latent variable, one attribute variable (for Multi-MNIST only), together with spatial attention variables, for efficient inference. We perform analysis to show that the learnt categorical distributions effectively capture the categories of objects in the scene for Multi-MNIST and for Multi-Sprites.

cs.CV↗

arXiv2017

Tactical Diagrammatic Reasoning

Although automated reasoning with diagrams has been possible for some years, tools for diagrammatic reasoning are generally much less sophisticated than their sentential cousins. The tasks of exploring levels of automation and abstraction in the construction of proofs and of providing explanations of solutions expressed in the proofs remain to be addressed. In this paper we take an interactive proof assistant for Euler diagrams, Speedith, and add tactics to its reasoning engine, providing a level of automation in the construction of proofs. By adding tactics to Speedith's repertoire of inferences, we ease the interaction between the user and the system and capture a higher level explanation of the essence of the proof. We analysed the design options for tactics by using metrics which relate to human readability, such as the number of inferences and the amount of clutter present in diagrams. Thus, in contrast to the normal case with sentential tactics, our tactics are designed to not only prove the theorem, but also to support explanation.

cs.LO↗