Papers
2021
From Poincaré Recurrence to Convergence in Imperfect Information Games: Finding Equilibrium via Regularization. J Perolat, R Munos, J-B Lespiau, S Omidshafiei, M Rowland, P Ortega, N Burch, TW Anthony, D Balduzzi, B De Vylder, G Piliouras, M Lanctot, K Tuyls
A Limited-Capacity Minimax Theorem for Non-Convex Games or: How I Learned to Stop Worrying about Mixed-Nash and Love Neural Nets. G Gidel, D Balduzzi, WM Czarnecki, M Garnelo, Y Bachrach
Pick Your Battles: Interaction Graphs as Population-Level Objectives for Strategic Diversity. M Garnelo, WM Czarnecki, S Liu, D Tirumala, J Oh, G Gidel, H van Hasselt, D Balduzzi
pdf, International Conference on Autonomous Agents and Multiagent Systems (AAMAS, extended abstract)
2020
From Chaos to Order: Symmetry and Conservation Laws in Game Dynamics. SG Nagarajan, D Balduzzi, G Piliouras
Real world games look like spinning tops. WM Czarnecki, G Gidel, B Tracey, K Tuyls, S Omidshafiei, D Balduzzi, M Jaderberg
Smooth markets: A basic mechanism for organizing gradient-based learners. D Balduzzi, WM Czarnecki, TW Anthony, IM Gemp, E Hughes, JZ Leibo, G Piliouras, T Graepel
Robust Self-organization in Games: Symmetries, Conservation Laws and Dimensionality Reduction. SG Nagarajan, D Balduzzi, G Piliouras
pdf, International Conference on Autonomous Agents and Multi-Agent Systems (AAMAS, extended abstract)
Learning to resolve alliance dilemmas in many-player zero-sum games. E Hughes, TW Anthony, T Eccles, JZ Leibo, D Balduzzi, Y Bachrach
D3C: Reducing the Price of Anarchy in Multi-Agent Learning. I Gemp, KR McKee, R Everett, EA Duéñez-Guzmán, Y Bachrach, D Balduzzi, A Tacchetti
2019
Open-ended learning in symmetric zero-sum games. D Balduzzi, M Garnelo, Y Bachrach, W Czarnecki, J Pérolat, M Jaderberg, T Graepel
Differentiable Game Mechanics. A Letcher*, D Balduzzi*, S Racanière, J Martens, J Foerster, K Tuyls, T Graepel
Stable Opponent Shaping in Differentiable Games. A Letcher, J Foerster, D Balduzzi, T Rocktäschel, S Whiteson
LOGAN: Latent Optimisation for Generative Adversarial Networks. Y Wu, J Donahue, D Balduzzi, K Simonyan, T Lillicrap
2018
Re-evaluating Evaluation. D Balduzzi, K Tuyls, J Pérolat, T Graepel
The Mechanics of n-Player Differentiable Games. D Balduzzi, S Racanière, J Martens, J Foerster, K Tuyls, T Graepel
pdf, slides, code, International Conference on Machine Learning (ICML).
[ best paper runner up at ICML ]
2017
The Shattered Gradients Problem: If resnets are the answer, then what is the question? D Balduzzi, M Frean, L Leary, JP Lewis, KW Ma, B McWilliams
pdf, slides, code for figure 1, video, International Conference on Machine Learning (ICML).
[ best paper award at Principled Approaches to Deep Learning workshop (ICML-PADL) ]
Neural Taylor Approximations: Convergence and Exploration in Rectifier Networks. D Balduzzi, B McWilliams, T Butler-Yeoman
Strongly-Typed Agents are Guaranteed to Interact Safely. D Balduzzi
Back to RGB: Deep Articulated Hand-Pose Estimation from a Single Camera Image. KW Ma, JP Lewis, M Frean, D Balduzzi
2016
Scatter Component Analysis: A Unified Framework for Domain Adaptation and Domain Generalization. M Ghifary, D Balduzzi, W Kleijn, M Zhang
Deep Reconstruction-Classification Networks for Unsupervised Domain Adaptation. M Ghifary, W Kleijn, M Zhang, D Balduzzi, W Li
Strongly-Typed Recurrent Neural Networks. D Balduzzi, M Ghifary
Compliance-aware bandits. N Della Penna, M Reid, D Balduzzi
pdf, Machine Learning for Healthcare Workshop (NeurIPS-ML4HC)
Grammars for Games: A Gradient-Based, Game-Theoretic Framework for Optimization in Deep Learning. D Balduzzi
Deep Online Convex Optimization with Gated Games. D Balduzzi
2015
Domain Generalization for Object Recognition with Multitask Autoencoders. M Ghifary, W Kleijn, M Zhang, D Balduzzi
Learning the Structure of Sum-Product Networks via an SVD-based Algorithm. T Adel, D Balduzzi, A Ghodsi
Kickback cuts Backprop's redtape: Biologically plausible credit assignment in neural networks. D Balduzzi, H Vanchinathan, J Buhmann
Compatible Value Gradients for Reinforcement Learning of Continuous Deep Policies. D Balduzzi, M Ghifary
2014
Uncovering the structure and temporal dynamics of information propagation. M Gomez Rodriguez, J Leskovec, D Balduzzi, B Schölkopf
Cortical prediction markets. D Balduzzi
Falsifiable implies Learnable. D Balduzzi
2013
Quantifying causal influences. D Janzing, D Balduzzi, M Grosse-Wentrup, B Schölkopf
Correlated random features for fast semi-supervised learning. B McWilliams, D Balduzzi, J Buhmann
Domain generalization via invariant feature representation. K Muandet, D Balduzzi, B Schölkopf
Randomized co-training: from cortical neurons to machine learning and back again. D Balduzzi
pdf, Randomized Methods for Machine Learning workshop (NeurIPS-RMML).
[ best paper award at NeurIPS-RMML ]
Pruning random features with correlated kitchen sinks (1 page abstract). B McWilliams, D Balduzzi
Metabolic cost as an organizing principle for cooperative learning. D Balduzzi, P Ortega, M Besserve
What can neurons do for their brain? Communicate selectivity with spikes. D Balduzzi, G Tononi
2012
Towards a learning-theoretic analysis of spike-timing dependent plasticity. D Balduzzi, M Besserve
A Nonparametric Conjugate Prior Distribution for the Maximizing Argument of a Noisy Function. P Ortega, J Grau-Moya, T Genewein, D Balduzzi, D Braun
Regulating the information in spikes: a useful bias. D Balduzzi
pdf, Information in Perception and Action workshop (NeurIPS-IPA)
A Neuromorphic Architecture for Object Recognition and Motion Anticipation Using Burst-STDP. A Nere, U Olcese, D Balduzzi, G Tononi
2011
Information, learning and falsification. D Balduzzi
pdf, slides, Philosophy and Machine Learning workshop (NeurIPS-PML)
Uncovering the Temporal Dynamics of Diffusion Networks. M Gomez Rodriguez, D Balduzzi, B Schölkopf
Falsification and future performance. D Balduzzi
pdf, Proceedings of Solomonoff 85th Memorial Conference, Lec Notes in Artificial Intelligence 7070, Springer.
On the information-theoretic structure of distributed measurements. D Balduzzi
pdf, dagstuhl, Elec Proc Theoretical Computer Science. 7th Developments of Computational Models workshop (DCM)
Estimating integrated information with TMS pulses during wakefulness, sleep, and under anesthesia. D Balduzzi
Detecting emergent processes in cellular automata with excess information. D Balduzzi
pdf, Advances in Artificial Life, MIT Press (ECAL)
2009
Qualia: The geometry of integrated information. D Balduzzi, G Tononi
pdf, PLoS Computational Biology 5(8): e1000462
Towards a theory of consciousness. G Tononi, D Balduzzi
pdf, slides, The Cognitive Neurosciences IV, edited by M Gazzaniga
2008
A BOLD window into brain waves. D Balduzzi, BA Riedner, G Tononi
Integrated Information in Discrete Dynamical Systems: Motivation and Theoretical Framework. D Balduzzi, G Tononi
pdf, code, PLoS Computational Biology 4(6): e1000091
Poisson geometry of parabolic bundles on elliptic curves. D Balduzzi
2006
Donagi-Markman cubic for Hitchin systems. D Balduzzi
Hamiltonian geometry of moduli space of bundles on curves. PhD thesis