Mónika Farsang

Mónika Farsang, Mathias Lechner, David Lung et al.

Bio-inspired neural networks have the potential to advance our understanding of neural computation and improve the state-of-the-art of AI systems. Bio-electrical synapses directly transmit neural signals, by enabling fast current flow between neurons. In contrast, bio-chemical synapses transmit neural signals indirectly, through neurotransmitters. Prior work showed that interpretable dynamics for complex robotic control, can be achieved by using chemical synapses, within a sparse, bio-inspired architecture, called Neural Circuit Policies (NCPs). However, a comparison of these two synaptic models, within the same architecture, remains an unexplored area. In this work we aim to determine the impact of using chemical synapses compared to electrical synapses, in both sparse and all-to-all connected networks. We conduct experiments with autonomous lane-keeping through a photorealistic autonomous driving simulator to evaluate their performance under diverse conditions and in the presence of noise. The experiments highlight the substantial influence of the architectural and synaptic-model choices, respectively. Our results show that employing chemical synapses yields noticeable improvements compared to electrical synapses, and that NCPs lead to better results in both synaptic models.

Mónika Farsang, Radu Grosu

In this paper, we present a unified framework for various bio-inspired models to better understand their structural and functional differences. We show that liquid-capacitance-extended models lead to interpretable behavior even in dense, all-to-all recurrent neural network (RNN) policies. We further demonstrate that incorporating chemical synapses improves interpretability and that combining chemical synapses with synaptic activation yields the most accurate and interpretable RNN models. To assess the accuracy and interpretability of these RNN policies, we consider the challenging lane-keeping control task and evaluate performance across multiple metrics, including turn-weighted validation loss, neural activity during driving, absolute correlation between neural activity and road trajectory, saliency maps of the networks' attention, and the robustness of their saliency maps measured by the structural similarity index.

Julian Lemmel, Felix Resch, Mónika Farsang et al.

Deploying pretrained policies in real-world applications presents substantial challenges that fundamentally limit the practical applicability of learning-based control systems. When autonomous systems encounter environmental changes in system dynamics, sensor drift, or task objectives, fixed policies rapidly degrade in performance. We show that employing Real-Time Recurrent Reinforcement Learning (RTRRL), a biologically plausible algorithm for online adaptation, can effectively fine-tune a pretrained policy to improve autonomous agents' performance on driving tasks. We further show that RTRRL synergizes with a recent biologically inspired recurrent network model, the Liquid-Resistance Liquid-Capacitance RNN. We demonstrate the effectiveness of this closed-loop approach in a simulated CarRacing environment and in a real-world line-following task with a RoboRacer car equipped with an event camera.

Mónika Farsang, Paul Mineiro, Wangda Zhang

Contextual bandits with average-case statistical guarantees are inadequate in risk-averse situations because they might trade off degraded worst-case behaviour for better average performance. Designing a risk-averse contextual bandit is challenging because exploration is necessary but risk-aversion is sensitive to the entire distribution of rewards; nonetheless we exhibit the first risk-averse contextual bandit algorithm with an online regret guarantee. We conduct experiments from diverse scenarios where worst-case outcomes should be avoided, from dynamic pricing, inventory management, and self-tuning software; including a production exascale data processing system.

Mónika Farsang, Ramin Hasani, Daniela Rus et al.

State Space Models (SSMs) are inherently recurrent along the sequence dimension, yet depth-recurrence - reusing the same block repeatedly across layers, as recently applied in looped transformers - has not been explored in this model family. We show that a looped SSM with $k$ parameters iterated $L$ times consistently closely matches or outperforms a standard SSM with $k \cdot L$ independent parameters across four architectures (LRU, S5, LinOSS, LrcSSM) and six time series classification benchmarks, despite operating within a strictly smaller hypothesis space, as we formally establish. Since the larger model contains the looped model as a special case, this dominance cannot be explained by expressivity and instead points to parameter sharing across depth as a beneficial inductive bias that simplifies optimization. These results demonstrate that depth-recurrence is orthogonal to sequence-recurrence and independently beneficial. We further show that input reshaping is an equally neglected design axis: concatenating timesteps for low-dimensional inputs, or flattening and rechunking the joint feature-time dimension for high-dimensional ones, yields accuracy gains of 1-6% across all models, confirmed over 5 random seeds. Both techniques provide standalone improvements that compound when combined, suggesting that depth and input reshaping are two independent and underexplored design axes for SSMs on time series.

Mónika Farsang, Ramin Hasani, Daniela Rus et al.

We present LrcSSM, a $\textit{non-linear}$ recurrent model that processes long sequences as fast as today's linear state-space layers. By forcing its Jacobian matrix to be diagonal, the full sequence can be solved in parallel, giving $\mathcal{O}(TD)$ time and memory and only $\mathcal{O}(\log T)$ sequential depth, for input-sequence length $T$ and a state dimension $D$. Moreover, LrcSSM offers a formal gradient-stability guarantee that other input-varying systems such as Liquid-S4 and Mamba do not provide. Importantly, the diagonal Jacobian structure of our model results in no performance loss compared to the original model with dense Jacobian, and the approach can be generalized to other non-linear recurrent models, demonstrating broader applicability. On a suite of long-range forecasting tasks, we demonstrate that LrcSSM outperforms Transformers, LRU, S5, and Mamba.

Elpiniki Maria Lygizou, Mónika Farsang, Radu Grosu

Transformers excel across a large variety of tasks but remain susceptible to corrupted inputs, since standard self-attention treats all query-key interactions uniformly. Inspired by lateral inhibition in biological neural circuits and building on the recent use by the Differential Transformer's use of two parallel softmax subtraction for noise cancellation, we propose Multihead Differential Gated Self-Attention (M-DGSA) that learns per-head input-dependent gating to dynamically suppress attention noise. Each head splits into excitatory and inhibitory branches whose dual softmax maps are fused by a sigmoid gate predicted from the token embedding, yielding a context-aware contrast enhancement. M-DGSA integrates seamlessly into existing Transformer stacks with minimal computational overhead. We evaluate on both vision and language benchmarks, demonstrating consistent robustness gains over vanilla Transformer, Vision Transformer, and Differential Transformer baselines. Our contributions are (i) a novel input-dependent gating mechanism for self-attention grounded in lateral inhibition, (ii) a principled synthesis of biological contrast-enhancement and self-attention theory, and (iii) comprehensive experiments demonstrating noise resilience and cross-domain applicability.

Mihaela-Larisa Clement, Mónika Farsang, Felix Resch et al.

Autonomous agents that rely purely on perception to make real-time control decisions require efficient and robust architectures. In this work, we demonstrate that augmenting RGB input with depth information significantly enhances our agents' ability to predict steering commands compared to using RGB alone. We benchmark lightweight recurrent controllers that leverage the fused RGB-D features for sequential decision-making. To train our models, we collect high-quality data using a small-scale autonomous car controlled by an expert driver via a physical steering wheel, capturing varying levels of steering difficulty. Our models were successfully deployed on real hardware and inherently avoided dynamic and static obstacles, under out-of-distribution conditions. Specifically, our findings reveal that the early fusion of depth data results in a highly robust controller, which remains effective even with frame drops and increased noise levels, without compromising the network's focus on the task.

Mónika Farsang, Luca Szegletes

Proximal Policy Optimization (PPO) is among the most widely used algorithms in reinforcement learning, which achieves state-of-the-art performance in many challenging problems. The keys to its success are the reliable policy updates through the clipping mechanism and the multiple epochs of minibatch updates. The aim of this research is to give new simple but effective alternatives to the former. For this, we propose linearly and exponentially decaying clipping range approaches throughout the training. With these, we would like to provide higher exploration at the beginning and stronger restrictions at the end of the learning phase. We investigate their performance in several classical control and locomotive robotic environments. During the analysis, we found that they influence the achieved rewards and are effective alternatives to the constant clipping method in many reinforcement learning tasks.

Mónika Farsang, Luca Szegletes

An in-depth understanding of the particular environment is crucial in reinforcement learning (RL). To address this challenge, the decision-making process of a mobile collaborative robotic assistant modeled by the Markov decision process (MDP) framework is studied in this paper. The optimal state-action combinations of the MDP are calculated with the non-linear Bellman optimality equations. This system of equations can be solved with relative ease by the computational power of Wolfram Mathematica, where the obtained optimal action-values point to the optimal policy. Unlike other RL algorithms, this methodology does not approximate the optimal behavior, it gives the exact, explicit solution, which provides a strong foundation for our study. With this, we offer new insights into understanding the action selection mechanisms in RL by presenting various small modifications on the very same schema that lead to different optimal policies.