Alessandro Ronca

Researcher in Artificial Intelligence

IRIS-AI

  • Machine Learning
  • Recurrent Neural Networks
  • Reinforcement Learning
  • Knowledge Representation
  • Automated Reasoning
  • Temporal Logics
Portrait of Alessandro Ronca

News

About

Alessandro Ronca is a researcher in Artificial Intelligence. In over 11 years of activity, he has led several research lines and made contributions across diverse areas of AI, including Deep Learning, particularly recurrent architectures; Reinforcement Learning, with an emphasis on history-dependent decision processes; and Knowledge Representation and Reasoning, especially temporal logics and stream reasoning.

His expertise bridges traditionally distinct areas of AI, from Machine Learning, grounded primarily in continuous mathematics and statistical inference, to Knowledge Representation and Reasoning, with its foundations in formal logic, discrete mathematics, and computation theory.

Alessandro is President and Co-Founder of IRIS-AI. He has been a Research Associate at the University of Oxford, and formerly at Sapienza University of Rome. He holds a PhD in Computer Science from the University of Oxford.

Research

Alessandro's research develops fundamental AI techniques for sequential data: time-series prediction, modelling token sequences, optimal decision making in the presence of temporal dependencies, and analytics over data streams.

Recurrent Neural Networks

Alessandro developed novel principles for analysing RNNs, which provide a better understanding of the capabilities and limitations of existing RNNs. This work led to MinMax Recurrent Neural Cascades, an architecture that addresses the identified limitations — it combines strong formal expressivity, efficient subquadratic inference, and provably stable dynamics with non-vanishing gradients.

  • Expressivity
  • Automata theory
  • Sequence models

Non-Markov Reinforcement Learning

Alessandro devised the first provably-optimal RL algorithms for Regular Decision Processes. They provide a principled foundation for agents having a guaranteed optimal behaviour, also in complex scenarios where the history of past events matters — and previously existing RL algorithms (e.g., for MDPs) would fail to achieve the same.

  • Regular decision processes
  • PAC learning
  • Offline RL

Temporal Logics

Alessandro developed Temporal Datalog, the first logic for reasoning over data streams with guarantees on the waiting times and memory requirements while avoiding the expressivity limitations imposed by sliding-window operators; and the Transformation Logics, which provide an unprecedented flexibility in matching expressivity requirements while keeping the computational cost of reasoning to a minimum.

  • Stream reasoning
  • Datalog
  • Complexity

Selected Publications

  1. NeurIPS2026
    MinMax Recurrent Neural CascadesAlessandro Ronca

    We introduce MinMax Recurrent Neural Cascades (MinMax RNCs), a class of recurrent neural networks based on a novel form of recurrence in the MinMax algebra. MinMax RNCs are a first in simultaneously enjoying the key properties for truly long-context processing: strong formal expressivity with long-term memory, efficient inference both in recurrent and parallel form, stable dynamics and gradients, and long-range sensitivity ensured by non-vanishing state gradients. First, they can store and retain information arbitrarily long, and manipulate it in a rich way, with a formal expressivity corresponding the regular languages—arguably the maximal expressivity for finite-memory systems. Second, in addition to linear-time inference via recurrence, they also admit log-depth–linear-work inference via parallel scan. Third, their states and activations are uniformly bounded independently of sequence length. Fourth, although not everywhere-differentiable, they admit pseudo-gradients that are uniformly bounded independently of sequence length and coincide with ordinary gradients almost everywhere. Fifth, they do not suffer from vanishing state gradients: the pseudo-gradient of a state with respect to an arbitrarily-distant past state can retain unit norm, while coinciding with the ordinary gradient. Empirically, we find that the theoretical properties translate to strong practical performance. We evaluate MinMax RNCs on a suite of fundamental synthetic tasks, and observe they solve all tasks perfectly, generalising seamlessly on longer sequences when tested up to length 1M, and outperforming a selection of baselines that well-represent the state of the art. We train a 112M-parameter MinMax RNC for next-token prediction in OWT, obtaining competitive performance for its size, and hence initial evidence of the potential of MinMax RNCs for real-world sequence-modelling tasks.

  2. NeurIPS2025
    Metric Automata Theory: A Unifying Theory of RNNsAdam Dankowiakowski, Alessandro Ronca

    We propose Metric Automata Theory, an elegant generalisation of classic Automata Theory to continuous dynamical systems, that constitutes a unifying theory of all kinds of Recurrent Neural Networks (RNNs), including widely-adopted architectures such as xLSTM and State Space Models (SSMs). The theory allows one to analyse RNNs both in the finite and unbounded precision settings seamlessly, while utilising fundamental results of Automata Theory. It also provides a novel notion of robustness that guarantees numerical stability and contributes to stability of learning. We employ the theory to prove a comprehensive set of expressivity results for widely-adopted RNNs, with a focus on robustness and finite-precision. Notably, we contrast the capabilities of xLSTM and SSMs for robustly modelling all star-free regular languages—xLSTM can do so, while SSMs cannot robustly recognize the FLIP-FLOP language. Thus we give a novel perspective on the importance of non-linear recurrences, giving insight for why xLSTM shows superior performance to SSMs on several tasks. We provide an improved understanding of the capabilities of Mamba, a popular SSM model. We show that Mamba is not generally capable of recognising the star-free languages under finite-precision, which is seemingly in contrast with the existing theoretical and empirical results for SSMs. We clarify the picture, by showing that Mamba admits a piecewise-linearly separable state space that allows it to approximate star-free languages, with some length-generalisation abilities. At the same time, Mamba does not admit such state spaces for languages like Parity. This explains why empirically Mamba performs well on star-free languages, and fails on Parity.

  3. ICLR2025
    Offline RL in Regular Decision Processes: Sample Efficiency via Language MetricsAhana Deb, Roberto Cipollone, Anders Jonsson, Alessandro Ronca, Mohammad Sadegh Talebi

    This work studies offline Reinforcement Learning (RL) in a class of non-Markovian environments called Regular Decision Processes (RDPs). In RDPs, the unknown dependency of future observations and rewards from the past interactions can be captured by some hidden finite-state automaton. For this reason, many RDP algorithms first reconstruct this unknown dependency using automata learning techniques. In this paper, we consider episodic RDPs and show that it is possible to overcome the limitations of existing offline RL algorithms for RDPs via the introduction of two original techniques: a novel metric grounded in formal language theory and an approach based on Count-Min-Sketch (CMS). Owing to the novel language metric, our algorithm is proven to be more sample efficient than existing results, and in some problem instances admitting low complexity languages, the gain is showcased to be exponential in the episode length. The CMS-based approach removes the need for naïve counting and alleviates the memory requirements for long planning horizons. We derive Probably Approximately Correct (PAC) sample complexity bounds associated to each of these techniques, and validate the approach experimentally.

  4. KR2024
    On the Expressivity of Recurrent Neural Cascades with IdentityNadezda Alexandrovna Knorozova, Alessandro Ronca
    PaperarXiv

    Recurrent Neural Cascades (RNC) are the class of recurrent neural networks with no cyclic dependencies among recurrent neurons. Their subclass RNC+ with positive recurrent weights has been shown to be closely connected to the star-free regular languages, which are the expressivity of many well-established temporal logics. The existing expressivity results show that the regular languages captured by RNC+ are the star-free ones, and they leave open the possibility that RNC+ may capture languages beyond regular. We exclude this possibility for languages that include an identity element, i.e., an input that can occur an arbitrary number of times without affecting the output. Namely, in the presence of an identity element, we show that the languages captured by RNC+ are exactly the star-free regular languages. Identity elements are ubiquitous in temporal patterns, and hence our results apply to a large number of applications. The implications of our results go beyond expressivity. At their core, we establish a close structural correspondence between RNC+ and semiautomata cascades, showing that every neuron can be equivalently captured by a three-state semiautomaton. A notable consequence of this result is that RNC+ are no more succinct than cascades of three-state semiautomata.

  5. AAAI2024
    On the Expressivity of Recurrent Neural CascadesNadezda Alexandrovna Knorozova, Alessandro Ronca
    PaperarXiv

    Recurrent Neural Cascades (RNCs) are the recurrent neural networks with no cyclic dependencies among recurrent neurons. This class of recurrent networks has received a lot of attention in practice. Besides training methods for a fixed architecture such as backpropagation, the cascade architecture naturally allows for constructive learning methods, where recurrent nodes are added incrementally one at a time, often yielding smaller networks. Furthermore, acyclicity amounts to a structural prior that even for the same number of neurons yields a more favourable sample complexity compared to a fully-connected architecture. A central question is whether the advantages of the cascade architecture come at the cost of a reduced expressivity. We provide new insights into this question. We show that the regular languages captured by RNCs with sign and tanh activation with positive recurrent weights are the star-free regular languages. In order to establish our results we developed a novel framework where capabilities of RNCs are accessed by analysing which semigroups and groups a single neuron is able to implement. A notable implication of our framework is that RNCs can achieve the expressivity of all regular languages by introducing neurons that can implement groups.

  6. IJCAI2024
    The Transformation LogicsAlessandro Ronca
    PaperarXiv

    We introduce a new family of temporal logics designed to finely balance the trade-off between expressivity and complexity. Their key feature is the possibility of defining operators of a new kind that we call transformation operators. Some of them subsume existing temporal operators, while others are entirely novel. Of particular interest are transformation operators based on semigroups. They enable logics to harness the richness of semigroup theory, and we show them to yield logics capable of creating hierarchies of increasing expressivity and complexity which are non-trivial to characterise in existing logics. The result is a genuinely novel and yet unexplored landscape of temporal logics, each of them with the potential of matching the trade-off between expressivity and complexity required by specific applications.

  7. NeurIPS2023
    Provably Efficient Offline Reinforcement Learning in Regular Decision ProcessesRoberto Cipollone, Anders Jonsson, Alessandro Ronca, Mohammad Sadegh Talebi

    This paper deals with offline (or batch) Reinforcement Learning (RL) in episodic Regular Decision Processes (RDPs). RDPs are the subclass of Non-Markov Decision Processes where the dependency on the history of past events can be captured by a finite-state automaton. We consider a setting where the automaton that underlies the RDP is unknown, and a learner strives to learn a near-optimal policy using pre-collected data, in the form of non-Markov sequences of observations, without further exploration. We present RegORL, an algorithm that suitably combines automata learning techniques and state-of-the-art algorithms for offline RL in MDPs. RegORL has a modular design allowing one to use any off-the-shelf offline RL algorithm in MDPs. We report a non-asymptotic high-probability sample complexity bound for RegORL to yield an ε-optimal policy, which makes appear a notion of concentrability relevant for RDPs. Furthermore, we present a sample complexity lower bound for offline RL in RDPs. To our best knowledge, this is the first work presenting a provably efficient algorithm for offline learning in RDPs.

  8. AAAI2023
    Automata Cascades: Expressivity and Sample ComplexityAlessandro Ronca, Nadezda Alexandrovna Knorozova, Giuseppe De Giacomo
    PaperarXiv

    Every automaton can be decomposed into a cascade of basic prime automata. This is the Prime Decomposition Theorem by Krohn and Rhodes. Guided by this theory, we propose automata cascades as a structured, modular, way to describe automata as complex systems made of many components, each implementing a specific functionality. Any automaton can serve as a component; using specific components allows for a fine-grained control of the expressivity of the resulting class of automata; using prime automata as components implies specific expressivity guarantees. Moreover, specifying automata as cascades allows for describing the sample complexity of automata in terms of their components. We show that the sample complexity is linear in the number of components and the maximum complexity of a single component, modulo logarithmic factors. This opens to the possibility of learning automata representing large dynamical systems consisting of many parts interacting with each other. It is in sharp contrast with the established understanding of the sample complexity of automata, described in terms of the overall number of states and input letters, which implies that it is only possible to learn automata where the number of states is linear in the amount of data available. Instead our results show that one can learn automata with a number of states that is exponential in the amount of data available.

  9. IJCAI2022
    Markov Abstractions for PAC Reinforcement Learning in Non-Markov Decision ProcessesAlessandro Ronca, Gabriel Paludo Licks, Giuseppe De Giacomo
    PaperarXiv

    Our work aims at developing reinforcement learning algorithms that do not rely on the Markov assumption. We consider the class of Non-Markov Decision Processes where histories can be abstracted into a finite set of states while preserving the dynamics. We call it a Markov abstraction since it induces a Markov Decision Process over a set of states that encode the non-Markov dynamics. This phenomenon underlies the recently introduced Regular Decision Processes (as well as POMDPs where only a finite number of belief states is reachable). In all such kinds of decision process, an agent that uses a Markov abstraction can rely on the Markov property to achieve optimal behaviour. We show that Markov abstractions can be learned during reinforcement learning. Our approach combines automata learning and classic reinforcement learning. For these two tasks, standard algorithms can be employed. We show that our approach has PAC guarantees when the employed algorithms have PAC guarantees, and we also provide an experimental evaluation.

  10. Artificial Intelligence2022
    The Delay and Window Size Problems in Rule-based Stream ReasoningAlessandro Ronca, Mark Kaminski, Bernardo Cuenca Grau, Ian Horrocks
    Paper

    In recent years, there has been an increasing interest in extending stream processing engines with rule-based temporal reasoning capabilities. To ensure correctness, such systems must be able to output results over the partial data received so far as if the entire (infinite) stream had been available; furthermore, these results must be streamed out as soon as the relevant data is received, thus incurring the minimum possible delay; finally, due to memory limitations, systems can only keep a limited history of previous facts in memory to perform further computations. These requirements pose significant theoretical and practical challenges since temporal rules can derive new information and propagate it both towards past and future time points; as a result, streamed answers can depend on data that has not yet been received, as well as on data that arrived far in the past. Towards developing a solid foundation for practical rule-based stream reasoning, we propose and study in this paper a suite of decision problems that can be exploited by stream reasoning algorithms to tackle the aforementioned challenges, and provide tight complexity bounds for a core temporal extension of Datalog. All of the problems we consider can be solved at design time (under reasonable assumptions), prior to the processing of any data. Solving these problems enables the use of reasoning algorithms that process the input streams incrementally using a sliding window, while at the same time supporting an expressive rule-based knowledge representation language and minimising both latency and memory consumption.

  11. IJCAI2021
    Efficient PAC Reinforcement Learning in Regular Decision ProcessesAlessandro Ronca, Giuseppe De Giacomo
    PaperarXiv

    Recently regular decision processes have been proposed as a well-behaved form of non-Markov decision process. Regular decision processes are characterised by a transition function and a reward function that depend on the whole history, though regularly (as in regular languages). In practice both the transition and the reward functions can be seen as finite transducers. We study reinforcement learning in regular decision processes. Our main contribution is to show that a near-optimal policy can be PAC-learned in polynomial time in a set of parameters that describe the underlying decision process. We argue that the identified set of parameters is minimal and it reasonably captures the difficulty of a regular decision process.

  12. KR2020
    Temporal Logic Monitoring Rewards via TransducersGiuseppe De Giacomo, Marco Favorito, Luca Iocchi, Fabio Patrizi, Alessandro Ronca
    Paper

    In Markov Decision Processes (MDPs), rewards are assigned according to a function of the last state and action. This is often limiting, when the considered domain is not naturally Markovian, but becomes so after careful engineering of extended state space. The extended states record information from the past that is sufficient to assign rewards by looking just at the last state and action. Non-Markovian Reward Decision Processes (NRMDPs) extend MDPs by allowing for non-Markovian rewards, which depend on the history of states and actions. Non-Markovian rewards can be specified in temporal logics on finite traces such as LTLf/LDLf, with the great advantage of a higher abstraction and succinctness; they can then be automatically compiled into an MDP with an extended state space. We contribute to the techniques to handle temporal rewards and to the solutions to engineer them. We first present an approach to compiling temporal rewards which merges the formula automata into a single transducer, sometimes saving up to an exponential number of states. We then define monitoring rewards, which add a further level of abstraction to temporal rewards by adopting the four-valued conditions of runtime monitoring; we argue that our compilation technique allows for an efficient handling of monitoring rewards. Finally, we discuss application to reinforcement learning.

  13. KR2018
    The Window Validity Problem in Rule-Based Stream ReasoningAlessandro Ronca, Mark Kaminski, Bernardo Cuenca Grau, Ian Horrocks
    PaperarXiv

    Rule-based temporal query languages provide the expressive power and flexibility required to capture in a natural way complex analysis tasks over streaming data. Stream processing applications, however, typically require near real-time response using limited resources. In particular, it becomes essential that the underpinning query language has favourable computational properties and that stream processing algorithms are able to keep only a small number of previously received facts in memory at any point in time without sacrificing correctness. In this paper, we propose a recursive fragment of temporal Datalog with tractable data complexity and study the properties of a generic stream reasoning algorithm for this fragment. We focus on the window validity problem as a way to minimise the number of time points for which the stream reasoning algorithm needs to keep data in memory at any point in time.

  14. AAAI2018
    Stream Reasoning in Temporal DatalogAlessandro Ronca, Mark Kaminski, Bernardo Cuenca Grau, Boris Motik, Ian Horrocks
    PaperarXiv

    In recent years, there has been an increasing interest in extending traditional stream processing engines with logical, rule-based, reasoning capabilities. This poses significant theoretical and practical challenges since rules can derive new information and propagate it both towards past and future time points; as a result, streamed query answers can depend on data that has not yet been received, as well as on data that arrived far in the past. Stream reasoning algorithms, however, must be able to stream out query answers as soon as possible, and can only keep a limited number of previous input facts in memory. In this paper, we propose novel reasoning problems to deal with these challenges, and study their computational properties on Datalog extended with a temporal sort and the successor function (a core rule-based language for stream reasoning applications).

All publications on Google Scholar →

Experience

Positions

Education

Service

Contact

For research collaborations, talks, or other enquiries, get in touch by email.