Abstract: A method includes generating, via a virtual machine, a group of code traces, each code trace of the group of code traces corresponding to a respective algorithm, of a group of algorithms, and a corresponding input. The method also includes fine-tuning a generative model in accordance with the group of code traces. The method further includes receiving, at the fine-tuned generative model, computer programming code. The method also includes generating, via the fine-tuned generative mode, one or more computer programming code statements corresponding to the computer programming code or simulate an expected output of the computer programming code.

Abstract: Certain aspects of the present disclosure provide techniques and apparatus for improved machine learning. In an example method, a current program state comprising a set of program instructions is accessed. A next program instruction is generated using a search operation, comprising generating a probability of the next program instruction based on processing the current program state and the next program instruction using a machine learning model, and generating a value of the next program instruction based on processing the current program state, the next program instruction, and a set of alternative outcomes using the machine learning model. An updated program state is generated based on adding the next program instruction to the set of program instructions.

Abstract: Certain aspects of the present disclosure provide techniques and apparatus for evaluating electronic circuit designs. A directed graph representing a netlist design for an electrical circuit is accessed, the netlist design comprising a plurality of electronic components and a plurality of connections among the plurality of electronic components. A node in the directed graph is selected, the node corresponding to a register that receives input from one or more of the plurality of electronic components in the netlist design. A subgraph is generated for the node, based on the directed graph, comprising identifying a connectivity cone ending at the first register. A functional embedding is generated for the subgraph based on a trained encoder machine learning model. A predicted performance characteristic of the netlist design is generated based at least in part on the functional embedding.