DreamCoder: growing generalizable, interpretable knowledge with wake-sleep Bayesian program learning.

Expert problem-solving is driven by powerful languages for thinking about problems and their solutions. Acquiring expertise means learning these languages-systems of concepts, alongside the skills to use them. We present DreamCoder, a system that learns to solve problems by writing programs. It builds expertise by creating domain-specific programming languages for expressing domain concepts, together with neural networks to guide the search for programs within these languages. A 'wake-sleep' learning algorithm alternately extends the language with new symbolic abstractions and trains the neural network on imagined and replayed problems. DreamCoder solves both classic inductive programming tasks and creative tasks such as drawing pictures and building scenes. It rediscovers the basics of modern functional programming, vector algebra and classical physics, including Newton's and Coulomb's laws. Concepts are built compositionally from those learned earlier, yielding multilayered symbolic representations that are interpretable and transferrable to new tasks, while still growing scalably and flexibly with experience. This article is part of a discussion meeting issue 'Cognitive artificial intelligence'.

Improved modeling of RNA-binding protein motifs in an interpretable neural model of RNA splicing.

Sequence-specific RNA-binding proteins (RBPs) play central roles in splicing decisions. Here, we describe a modular splicing architecture that leverages in vitro-derived RNA affinity models for 79 human RBPs and the annotated human genome to produce improved models of RBP binding and activity. Binding and activity are modeled by separate Motif and Aggregator components that can be mixed and matched, enforcing sparsity to improve interpretability. Training a new Adjusted Motif (AM) architecture on the splicing task not only yields better splicing predictions but also improves prediction of RBP-binding sites in vivo and of splicing activity, assessed using independent data.

Predicting response to a smartphone-based cognitive-behavioral therapy for body dysmorphic disorder.

BACKGROUND: Body dysmorphic disorder (BDD) is a severe, chronic disorder if untreated. Smartphone cognitive behavioral therapy (CBT) for BDD is efficacious and can reduce key treatment barriers (e.g., lack of clinicians, cost, stigma). While promising, little is known about who is more or less likely to benefit from this approach. METHODS: This is a secondary data analysis of a randomized, waitlist-controlled trial of smartphone CBT for BDD. Participants (N = 80) were recruited nationally and randomized to receive a 12-week, coach-guided CBT for BDD app, either immediately or after a 12-week waitlist. The main outcome for this analysis was BDD severity (BDD-YBOCS) over time (baseline, week 6, week 12) during the active app use phase in each randomized group (n = 74). Secondary outcomes included treatment response (≥30 % reduction in BDD-YBOCS) and remission (total BDD-YBOCS ≤16) at end-of-treatment. RESULTS: Immediate (vs. delayed) CBT predicted better outcomes (symptom improvement), as did gender identity (symptom improvement), higher baseline treatment credibility and expectancy (response, remission), lower baseline BDD severity (remission), and sexual minority status (vs. heterosexual; response, remission). LIMITATIONS: Limitations include the relatively small sample, drop-out rate of 22 %, and limited gender and racial-ethnic diversity. CONCLUSIONS: These results highlight a potential advantage of smartphone CBT in historically marginalized populations, and the importance of efforts to hasten treatment access, bolster confidence in the treatment at treatment onset, and develop stratified care models to optimize treatment allocation and efficacy.

Synthesizing theories of human language with Bayesian program induction.

Automated, data-driven construction and evaluation of scientific models and theories is a long-standing challenge in artificial intelligence. We present a framework for algorithmically synthesizing models of a basic part of human language: morpho-phonology, the system that builds word forms from sounds. We integrate Bayesian inference with program synthesis and representations inspired by linguistic theory and cognitive models of learning and discovery. Across 70 datasets from 58 diverse languages, our system synthesizes human-interpretable models for core aspects of each language's morpho-phonology, sometimes approaching models posited by human linguists. Joint inference across all 70 data sets automatically synthesizes a meta-model encoding interpretable cross-language typological tendencies. Finally, the same algorithm captures few-shot learning dynamics, acquiring new morphophonological rules from just one or a few examples. These results suggest routes to more powerful machine-enabled discovery of interpretable models in linguistics and other scientific domains.