382 citations · 916 across the 31 of their papers we have counts for
32 papers
DiffusionGemma Technical Report
DiffusionGemma Team, Adrien Ali Taïga, James Assiene +41
We introduce DiffusionGemma, an experimental open-weight language model that uses discrete diffusion to generate text at exceptionally high speed. Rather than decoding one token at…
How Transparent is DiffusionGemma?
Joshua Engels, Callum McDougall, Bilal Chughtai +11
LLM reasoning transparency is a critical affordance for understanding model decisions, mitigating misuse and misalignment, and debugging surprising model behaviors. However, Diffus…
PDLP: A Practical First-Order Method for Large-Scale Linear Programming
David Applegate, Mateo Díaz, Oliver Hinder +4
We present PDLP, a practical first-order method for linear programming (LP) designed to solve large-scale LP problems. PDLP is based on the primal-dual hybrid gradient (PDHG) metho…
Probabilistic Inference in Reinforcement Learning Done Right
Jean Tarbouriech, Tor Lattimore, Brendan O'Donoghue
A popular perspective in Reinforcement learning (RL) casts the problem as probabilistic inference on a graphical model of the Markov decision process (MDP). The core object of stud…
ReLOAD: Reinforcement Learning with Optimistic Ascent-Descent for Last-Iterate Convergence in Constrained MDPs
Ted Moskovitz, Brendan O'Donoghue, Vivek Veeriah +3
In recent years, Reinforcement Learning (RL) has been applied to real-world problems with increasing success. Such applications often require to put constraints on the agent's beha…
Efficient Exploration via Epistemic-Risk-Seeking Policy Optimization
Brendan O'Donoghue
Exploration remains a key challenge in deep reinforcement learning (RL). Optimism in the face of uncertainty is a well-known heuristic with theoretical guarantees in the tabular se…
Optimistic Meta-Gradients
Sebastian Flennerhag, Tom Zahavy, Brendan O'Donoghue +3
We study the connection between gradient-based meta-learning and convex op-timisation. We observe that gradient descent with momentum is a special case of meta-gradients, and build…
POMRL: No-Regret Learning-to-Plan with Increasing Horizons
Khimya Khetarpal, Claire Vernade, Brendan O'Donoghue +2
We study the problem of planning under model uncertainty in an online meta-reinforcement learning (RL) setting where an agent is presented with a sequence of related tasks with lim…
On the connection between Bregman divergence and value in regularized Markov decision processes
Brendan O'Donoghue
In this short note we derive a relationship between the Bregman divergence from the current policy to the optimal policy and the suboptimality of the current value function in a re…
Variational Bayesian Optimistic Sampling
Brendan O'Donoghue, Tor Lattimore
We consider online sequential decision problems where an agent must balance exploration and exploitation. We derive a set of Bayesian `optimistic' policies which, in the stochastic…
The Neural Testbed: Evaluating Joint Predictions
Ian Osband, Zheng Wen, Seyed Mohammad Asghari +7
Predictive distributions quantify uncertainties ignored by point estimates. This paper introduces The Neural Testbed: an open-source benchmark for controlled and principled evaluat…
Practical Large-Scale Linear Programming using Primal-Dual Hybrid Gradient
David Applegate, Mateo Díaz, Oliver Hinder +4
We present PDLP, a practical first-order method for linear programming (LP) that can solve to the high levels of accuracy that are expected in traditional LP applications. In addit…
Discovering Diverse Nearly Optimal Policies with Successor Features
Tom Zahavy, Brendan O'Donoghue, Andre Barreto +3
Finding different solutions to the same problem is a key aspect of intelligence associated with creativity and adaptation to novel situations. In reinforcement learning, a set of d…
Reward is enough for convex MDPs
Tom Zahavy, Brendan O'Donoghue, Guillaume Desjardins +1
Maximising a cumulative reward function that is Markov and stationary, i.e., defined over state-action pairs and independent of time, is sufficient to capture many kinds of goals i…
Discovering a set of policies for the worst case reward
Tom Zahavy, Andre Barreto, Daniel J Mankowitz +4
We study the problem of how to construct a set of policies that can be composed together to solve a collection of reinforcement learning tasks. Each task is a different reward func…
Solving Mixed Integer Programs Using Neural Networks
Vinod Nair, Sergey Bartunov, Felix Gimeno +16
Mixed Integer Programming (MIP) solvers rely on an array of sophisticated heuristics developed with decades of research to solve large-scale MIP instances encountered in practice.…
Sample Efficient Reinforcement Learning with REINFORCE
Junzi Zhang, Jongho Kim, Brendan O'Donoghue +1
Policy gradient methods are among the most effective methods for large-scale reinforcement learning, and their empirical success has prompted several works that develop the foundat…
Matrix games with bandit feedback
Brendan O'Donoghue, Tor Lattimore, Ian Osband
We study a version of the classical zero-sum matrix game with unknown payoff matrix and bandit feedback, where the players only observe each others actions and a noisy payoff. This…
Operator splitting for a homogeneous embedding of the linear complementarity problem
Brendan O'Donoghue
The linear complementarity problem (LCP) is a general set membership problem that includes quadratic cone programming as a special case. In this work we consider a homogeneous embe…
Making Sense of Reinforcement Learning and Probabilistic Inference
Brendan O'Donoghue, Ian Osband, Catalin Ionescu
Reinforcement learning (RL) combines a control problem with statistical estimation: The system dynamics are not known to the agent, but can be learned through experience. A recent…
Hamiltonian descent for composite objectives
Brendan O'Donoghue, Chris J. Maddison
In optimization the duality gap between the primal and the dual problems is a measure of the suboptimality of any primal-dual point. In classical mechanics the equations of motion…
Verification of Non-Linear Specifications for Neural Networks
Chongli Qin, Krishnamurthy, Dvijotham +7
Prior work on neural network verification has focused on specifications that are linear functions of the output of the network, e.g., invariance of the classifier output under adve…
Strength in Numbers: Trading-off Robustness and Computation via Adversarially-Trained Ensembles
Edward Grefenstette, Robert Stanforth, Brendan O'Donoghue +3
While deep learning has led to remarkable results on a number of challenging problems, researchers have discovered a vulnerability of neural networks in adversarial settings, where…
Globally Convergent Type-I Anderson Acceleration for Non-Smooth Fixed-Point Iterations
Junzi Zhang, Brendan O'Donoghue, Stephen Boyd
We consider the application of the type-I Anderson acceleration to solving general non-smooth fixed-point problems. By interleaving with safe-guarding steps, and employing a Powell…