Online Streaming Features Selection via Markov Blanket

Streaming feature selection has always been an excellent method for selecting the relevant subset of features from high-dimensional data and overcoming learning complexity. However, little attention is paid to online feature selection through the Markov Blanket (MB). Several studies based on traditional MB learning presented low prediction accuracy and used fewer datasets as the number of conditional independence tests is high and consumes more time. This paper presents a novel algorithm called Online Feature Selection Via Markov Blanket (OFSVMB) based on a statistical conditional independence test offering high accuracy and less computation time. It reduces the number of conditional independence tests and incorporates the online relevance and redundant analysis to check the relevancy between the upcoming feature and target variable T, discard the redundant features from Parents-Child (PC) and Spouses (SP) online, and find PC and SP simultaneously. The performance OFSVMB is compared with traditional MB learning algorithms including IAMB, STMB, HITON-MB, BAMB, and EEMB, and Streaming feature selection algorithms including OSFS, Alpha-investing, and SAOLA on 9 benchmark Bayesian Network (BN) datasets and 14 real-world datasets. For the performance evaluation, F1, precision, and recall measures are used with a significant level of 0.01 and 0.05 on benchmark BN and real-world datasets, including 12 classifiers keeping a significant level of 0.01. On benchmark BN datasets with 500 and 5000 sample sizes, OFSVMB achieved significant accuracy than IAMB, STMB, HITON-MB, BAMB, and EEMB in terms of F1, precision, recall, and running faster. It finds more accurate MB regardless of the size of the features set. In contrast, OFSVMB offers substantial improvements based on mean prediction accuracy regarding 12 classifiers with small and large sample sizes on real-world datasets than OSFS, Alpha-investing, and SAOLA but slower than OSFS, Alpha-investing, and SAOLA because these algorithms only find the PC set but not SP. Furthermore, the sensitivity analysis shows that OFSVMB is more accurate in selecting the optimal features.

Recurrent, nonequilibrium systems and the Markov blanket assumption

Markov blankets –statistical independences between system and environment– have become popular to describe the boundaries of living systems under Bayesian views of cognition. The intuition behind Markov blanket originates from considering acyclic, atemporal networks. In contrast, living systems display recurrent interactions that generate pervasive couplings between system and environment, making Markov blankets highly unusual and restricted to particular cases.

Who tailors the blanket?

The gap between the Markov blanket and ontological boundaries arises from the former’s inability to capture the dynamic process through which biological and cognitive agents actively generate their own boundaries with the environment. Active inference in the FEP framework presupposes the existence of a Markov blanket, but it is not a process that actively generates the latter.

Does the metaphysical dog wag its formal tail? The free energy principle and philosophical debates about life, mind, and matter.

According to Bruineberg and colleagues, philosophical arguments on life, mind, and matter that are based on the free energy principle (FEP) (i) essentially draw on the Markov blanket construct and (ii) tend to assume that strong metaphysical claims can be justified on the basis of metaphysically innocuous formal assumptions provided by FEP. I argue against both (i) and (ii).

The Seductive Allure of Cargo Cult Computationalism

Bruineberg and colleagues report a striking confusion, in which the formal Bayesian notion of a “Markov Blanket” has been frequently misunderstood and misapplied to phenomena of mind and life. I argue that misappropriation of formal concepts is pervasive in the “predictive processing” literature, and echo Richard Feynman in suggesting how we might resist the allure of cargo cult computationalism.

Causal surgery under a Markov blanket

Bruineberg et al provide compelling clarity on the roles Markov blankets could (and perhaps should) play in the study of life and mind. However, here we draw attention to a further role blankets might play: as a hypothesis about cognition itself. People and other animals may use blanket-like representations to model the boundary between themselves and their worlds.

Emergent classicality in general multipartite states and channels

In a quantum measurement process, classical information about the measured system spreads throughout the environment. Meanwhile, quantum information about the system becomes inaccessible to local observers. Here we prove a result about quantum channels indicating that an aspect of this phenomenon is completely general. We show that for any evolution of the system and environment, for everywhere in the environment excluding an O(1)-sized region we call the "quantum Markov blanket," any locally accessible information about the system must be approximately classical, i.e. obtainable from some fixed measurement. The result strengthens the earlier result of Brandão et al. (Nat. comm. 6:7908) in which the excluded region was allowed to grow with total environment size. It may also be seen as a new consequence of the principles of no-cloning or monogamy of entanglement. Our proof offers a constructive optimization procedure for determining the "quantum Markov blanket" region, as well as the effective measurement induced by the evolution. Alternatively, under channel-state duality, our result characterizes the marginals of multipartite states.