Multi-Label Classification Neural Networks with Hard Logical Constraints

Multi-label classification (MC) is a standard machine learning problem in which a data point can be associated with a set of classes. A more challenging scenario is given by hierarchical multi-label classification (HMC) problems, in which every prediction must satisfy a given set of hard constraints expressing subclass relationships between classes. In this article, we propose C-HMCNN(h), a novel approach for solving HMC problems, which, given a network h for the underlying MC problem, exploits the hierarchy information in order to produce predictions coherent with the constraints and to improve performance. Furthermore, we extend the logic used to express HMC constraints in order to be able to specify more complex relations among the classes and propose a new model CCN(h), which extends C-HMCNN(h) and is again able to satisfy and exploit the constraints to improve performance. We conduct an extensive experimental analysis showing the superior performance of both C-HMCNN(h) and CCN(h) when compared to state-of-the-art models in both the HMC and the general MC setting with hard logical constraints.

Bits and pieces: understanding information decomposition from part-whole relationships and formal logic

Partial information decomposition (PID) seeks to decompose the multivariate mutual information that a set of source variables contains about a target variable into basic pieces, the so-called ‘atoms of information’. Each atom describes a distinct way in which the sources may contain information about the target. For instance, some information may be contained uniquely in a particular source, some information may be shared by multiple sources and some information may only become accessible synergistically if multiple sources are combined. In this paper, we show that the entire theory of PID can be derived, firstly, from considerations of part-whole relationships between information atoms and mutual information terms, and secondly, based on a hierarchy of logical constraints describing how a given information atom can be accessed. In this way, the idea of a PID is developed on the basis of two of the most elementary relationships in nature: the part-whole relationship and the relation of logical implication. This unifying perspective provides insights into pressing questions in the field such as the possibility of constructing a PID based on concepts other than redundant information in the general n-sources case. Additionally, it admits of a particularly accessible exposition of PID theory.

Causality-Based Game Solving

We present a causality-based algorithm for solving two-player reachability games represented by logical constraints. These games are a useful formalism to model a wide array of problems arising, e.g., in program synthesis. Our technique for solving these games is based on the notion of subgoals, which are slices of the game that the reachability player necessarily needs to pass through in order to reach the goal. We use Craig interpolation to identify these necessary sets of moves and recursively slice the game along these subgoals. Our approach allows us to infer winning strategies that are structured along the subgoals. If the game is won by the reachability player, this is a strategy that progresses through the subgoals towards the final goal; if the game is won by the safety player, it is a permissive strategy that completely avoids a single subgoal. We evaluate our prototype implementation on a range of different games. On multiple benchmark families, our prototype scales dramatically better than previously available tools.

Could God Love Cruelty?

One of the foremost objections to theological voluntarism is the contingency objection. If God’s will fixes moral facts, then what if God willed that agents engage in cruelty? I argue that even unrestricted theological voluntarists should accept some logical constraints on possible moral systems—hence, some limits on ways that God could have willed morality to be—and these logical constraints are sufficient to blunt the force of the contingency objection. One constraint I defend is a very weak accessibility requirement, related to (but less problematic than) existence internalism in metaethics. The theological voluntarist can maintain: Godcouldn’t have loved cruelty, and even though he could have willed behaviors we find abhorrent, he could only have done so in a world of deeply alien moral agents. We cannot confidently declare such a world unacceptable.

THEORETICAL DEFINITIONS FOR AUXILIARY CONCEPTS OF VARIOUS LEVELS OF ABSTRACTION

Theoretical knowledge may contain various levels of abstraction represented by logical constructions from the observed characteristics of objects from the subject area of the theory. The degree of abstractness can be de-scribed by the complexity of the structures obtained from the initial observational terms. Such auxiliary construc-tions are characterized as explicit or implicitdefinitions of theoretical concepts in terms of observational. One of the techniques for constructing such definitions is the operationalization of abstractions by a system of reduction sentences. In this case a theoretical concept is characterized as “open” and plays a role of logical and methodo-logical constraints for expanding the possible connections of the theoretical model with the help of concepts from the domain of intended practical application.

Knowledge of uncertain worlds: programming with logical constraints

Programming with logic for sophisticated applications must deal with recursion and negation, which together have created significant challenges in logic, leading to many different, conflicting semantics of rules. This paper describes a unified language, DA logic, for design and analysis logic, based on the unifying founded semantics and constraint semantics, that supports the power and ease of programming with different intended semantics. The key idea is to provide meta-constraints, support the use of uncertain information in the form of either undefined values or possible combinations of values and promote the use of knowledge units that can be instantiated by any new predicates, including predicates with additional arguments.

An Exact Algorithm for the Optimal Chiller Loading Problem and Its Application to the Optimal Chiller Sequencing Problem

The optimal management of multiple chiller systems calls for the solution of the so-called optimal chiller loading (OCL) problem. Due to the interplay of continuous and logical constraints, OCL is an NP-hard problem, so that a variety of heuristic algorithms have been proposed in the literature. Herein, an algorithm for its exact solution, named X-OCL, is developed under the assumption that the chillers’ power consumption curves are quadratic. The proposed method hinges on a decomposition of the solution space so that the overall OCL problem is decomposed to a set of equality constrained quadratic programming problems that can be solved in closed form. By applying the new X-OCL solver to well known case studies, we assess and compare the performances of several literature algorithms, highlighting also some errors in the published results. Moreover, X-OCL is used to design a greedy optimal chiller sequencing (OCS) solver, called X-OCS. The X-OCS is tested on two literature benchmarks and on the model of the heating, ventilation and air-conditioning (HVAC) system of a semiconductor plant, over a two-year period. The performances of X-OCS are remarkably close to the theoretical optimal performance.

Learning Probabilistic Sentential Decision Diagrams by Sampling

Probabilistic circuits are deep probabilistic models with neural-network-like semantics capable of accurately and efficiently answering probabilistic queries without sacrificing expressiveness. Probabilistic Sentential Decision Diagrams (PSDDs) are a subclass of probabilistic circuits able to embed logical constraints to the circuit’s structure. In doing so, they obtain extra expressiveness with empirical optimal performance. Despite achieving competitive performance compared to other state-of-the-art competitors, there have been very few attempts at learning PSDDs from a combination of both data and knowledge in the form of logical formulae. Our work investigates sampling random PSDDs consistent with domain knowledge and evaluating against state-of-the-art probabilistic models. We propose a method of sampling that retains important structural constraints on the circuit’s graph that guarantee query tractability. Finally, we show that these samples are able to achieve competitive performance even on larger domains.