Abstract: An analog machine learning architecture uses modular analog multiplier-accumulator (AMAC) elements of fixed size to form a machine learning (ML) system with increasing feature map size. A single 3 × 3 × 64 AMAC array is arranged to provide a three layer ML architecture with first layer 3×3×64, second layer 3×3×128, and third layer 3×3×256 using arrangements of single 3×3×64 AMACs arranged in parallel, where the bias of each AMAC is separately established in a unique interval of time.

Abstract: A lightweight node in a decentralized network includes stores a blockchain with a plurality of blocks. The lightweight node adds blocks to the blockchain successively. A given block having a header and a body. The header includes a data merkle root generated as a root hash of a data merkle tree with one or more leaf nodes that are one or more hashes. A given hash being a hash of a combination of (1) a public key associated with a lightweight node of the decentralized network and (2) of a validity value associated with the public key indicating whether the public key is a valid public key. The data merkle root being insufficient for restoring the data merkle tree. But with a public key and an intermediate hash the date merkle root is sufficient for at least partly verifying the public key.

Abstract: An unenrolled lightweight node is on a decentralized network with a trusted node and a plurality of peers. The unenrolled lightweight node and the peers run a lightweight blockchain consensus algorithm. The unenrolled lightweight node includes (a) circuitry for storing a token that includes a signature that includes at least a signature of at least a first identifier signed with a private key of the trusted node, the first identifier being associated with a public key of the unenrolled lightweight node, and (b) circuitry for broadcasting a request for blockchain enrollment of the unenrolled lightweight node to the plurality of peers. The authentication request including at least a second identifier that is associated with at least a public key of the unenrolled lightweight node, a signature created with at least the second identifier and a corresponding private key of the unenrolled lightweight node, and the token.

Abstract: A process for a floating point multiplier-accumulator (MAC) is operative on N pairs of floating point values using N MAC processes operating concurrently, each MAC process operating on a pair of values comprising an input value and a coefficient value. Each MAC process simultaneously generates: an integer form fraction at a first bitwidth and a second bitwidth greater than the first bitwidth, a sign bit, and an exponent difference computed by subtracting an exponent sum from a maximum exponent sum of all exponent sums. The integer form fractions of the first bitwidths are provided to an adder tree using the first bitwidth, and if the sum has an excess percentage of leading 0s, then the second bitwidth is used by an adder tree using the second bitwidth to form a great precision integer form fraction. The sign, integer form fraction, and maximum exponent are provided to an normalizer which generates a floating point result.

Abstract: A floating point multiplier-accumulator (MAC) multiplies and accumulates N pairs of floating point values using N MAC processors operating simultaneously, each pair of values comprising an input value and a coefficient value to be multiplied and accumulated. The pairs of floating point values are simultaneously processed by the plurality of MAC processors, each of which output a signed integer form fraction with a first bitwidth and a second bitwith, along with a maximum exponent. The first bitwidth signed integer form fractions are summed by an adder tree using the first bitwidth to form a first sum, and when an excess leading 0 condition is detected, a second adder tree operative on the second bitwidth integer form fractions forms a second sum. The first sum or second sum, along with the maximum exponent, is converted into floating point result.

Abstract: A process for a floating point multiplier-accumulator (MAC) is operative on N pairs of floating point values using N MAC processes operating concurrently, each MAC process operating on a pair of values comprising an input value and a coefficient value. Each MAC process simultaneously generates an integer form fraction accompanied by a sign bit and an exponent difference computed by subtracting an exponent sum from a maximum exponent sum of all exponent sums. A range estimating process determines a possible range of values from the exponent differences and determines an adder precision. A summing process adds all of the integer form fractions using the determined adder precision, and converts the sum to a floating point value using the maximum exponent sum, sign bit of the summed integer form fractions, and optionally performs a 2's complement of the summed integer form fraction if the sign bit is negative.

Abstract: A floating point multiplier-accumulator (MAC) multiplies and accumulates N pairs of floating point values using N MAC processors operating simultaneously, each pair of values comprising an input value and a coefficient value to be multiplied and accumulated. The pairs of floating point values are simultaneously processed by the plurality of MAC processors, each of which outputs a signed integer form fraction and a maximum exponent. A range estimator forms a possible range of values from the exponent differences and determines an adder precision. The integer form fractions are summed using the adder precision, a sign bit is extracted, and a floating point value is output. Each MAC processor provides its integer form fraction with a precision determined by the MAC processor's exponent difference.

Abstract: A node mesh contains an originating node coupled to one or more nodes, each node having an communications interface input and a communications interface output. Each node has a route table with an association between a header amplitude and an output interface, such that a header having a particular amplitude causes the input node which received the message to couple the message to an associated communications interface output of the node. When the originating node outputs a message with a header amplitude, each node of the node mesh in turn directs the message to an output interface as directed by the node local route table to a terminating node of the node mesh, where the terminating node may be a training processor or inference processor for machine learning applications.

Abstract: A Bias Unit Element (UE) has a digital input and sign input, and comprises a positive Bias UE and a negative Bias UE, each comprising groups of NAND gates generating an output and a complementary output, each of which are coupled to differential charge transfer lines through binary weighted charge transfer capacitors to a differential charge transfer bus comprising a positive charge transfer line and a negative charge transfer line. The sign input enables the positive Bias UE when the sign bit is positive and enables the negative Bias UE when the sign bit is negative.

Abstract: A Unit Element (UE) has a digital X input and a digital W input, and comprises groups of NAND gates generating complementary outputs which are coupled to a differential charge transfer bus comprising a positive charge transfer line and a negative charge transfer line. The number of bits in the X input determines the number of NAND gates in a NAND-group and the number of bits in the W input determines the number of NAND groups. Each NAND-group receives one bit of the W input applied to all of the NAND gates of the NAND-group, and each unit element having the bits of X applied to each associated NAND gate input of each unit element. The NAND gate outputs are coupled through binary weighted charge transfer capacitors to a positive charge transfer line and negative charge transfer line.

Abstract: An architecture for a chopper stabilized multiplier-accumulator (MAC) uses a chop clock and common Unit Element (UE), the MAC formed as a plurality of MAC UEs receiving X and W values and a sign bit exclusive ORed with the chop clock, a plurality of Bias UEs receiving E value and a sign bit exclusive ORed with the chop clock, and a plurality of Analog to Digital Conversion (ADC) UEs which collectively perform a scalable MAC operation and generate a binary result. Each MAC UE, BIAS UE and ADC UE comprises groups of NAND gates with complementary outputs arranged in NAND-groups, each NAND gate coupled to a differential charge transfer bus through a binary weighted charge transfer capacitor. The analog charge transfer bus is coupled to groups of ADC UEs with an ADC controller which enables and disables the ADC UEs using successive approximation to determine the accumulated multiplication result.

Abstract: A Bias Unit Element (UE) has a digital input, a sign input, and a chop clock. The sign input is exclusive ORed with the chop clock to generate a signed chop clock. Each Bias UE comprises a positive Bias UE and a negative Bias UE, each comprising groups of NAND gates generating an output and a complementary output, each of which are coupled to differential charge transfer lines through binary weighted charge transfer capacitors to a differential charge transfer bus comprising a positive charge transfer line and a negative charge transfer line. The chopped sign input enables the positive Bias UE when the sign bit is positive and enables the negative Bias UE when the sign bit is negative.

Abstract: A Unit Element (UE) has a positive UE and a negative UE, each having a digital X input and a digital W input with a sign bit, the sign bit is exclusive ORed with a chop clock to generate a chopped sign bit. The positive UE is enabled when the chopped sign bit is positive and the negative UE is enabled when the chopped sign bit is negative. Each positive and negative UE comprises groups of NAND gates generating an output and complementary output which are coupled to a differential charge transfer bus comprising a positive charge transfer line and a negative charge transfer line. The NAND gate outputs and complementary outputs are coupled through binary weighted charge transfer capacitors the positive charge transfer line and negative charge transfer line.

Abstract: A Gain Balanced Analog Multiply-Accumulator (AMAC) has an inference memory which outputs subsets of inference data comprising X input values and one or more associated W coefficient values. The Gain Balanced AMAC has a number of Analog Multiplier-Accumulator Unit Elements (AMAC UE) in equal number to the number of X input values in each subset of inference data. In each of a series of multiply-accumulate cycles, the X input values and one or more W coefficient values from the inference memory are applied to each AMAC UE to generate a charge corresponding to the multiplication of X input value and W coefficient value of each AMAC UE which is transferred to a shared analog charge bus. The inference memory applies the X input value and W coefficient values of each subset to a different AMAC UE on subsequent cycles to balance the gain of the AMAC such that gain differences from one AMAC UE to another are not cumulative.

Abstract: A node mesh contains an originating node and several node groups, each node group consisting of one or more nodes with interfaces connected to other nodes of the node group. Each node of a node group has an associated route table with an association between an applied DC voltage and an output interface to couple the input signal to. When the originating node outputs a DC voltage accompanied by differential signaling, each node in turn directs the DC voltage and differential signaling to an output interface as directed by the node local route table to a local termination in a node, which may be coupled to a training processor of inference processor for machine learning applications.

Abstract: An architecture for a multiplier-accumulator (MAC) uses a common Unit Element (UE) for each aspect of operation, the MAC formed as a plurality of MAC UEs, a plurality of Bias UEs, and a plurality of Analog to Digital Conversion (ADC) UEs which collectively perform a scalable MAC operation and generate a binary result. Each MAC UE, BIAS UE and ADC UE comprises groups of NAND gates with complementary outputs arranged in NAND-groups, each NAND gate coupled to a differential charge transfer bus through a binary weighted charge transfer capacitor to form an analog multiplication product as a charge applied to the differential charge transfer bus. The analog charge transfer bus is coupled to groups of ADC UEs with an ADC controller which enables and disables the ADC UEs using successive approximation to determine the accumulated multiplication result.

Abstract: A multiplier is formed from a plurality of nanomagnetic structures including slant edge input nanomagnetic structures, diagonal elongate interconnect nanomagnetic structures, and output nanomagnetic structures. Input logic levels are provided by inserting a magnetic field, which generates an set of output magnetic fields representing the product of the binary input values.

Abstract: A multiplier device for binary magnetic applied fields uses Interlayer Exchange Coupling (IEC) structure where two layers of ferromagnetic material are separated from each other by non-magnetic layers of electrically conductive material of atomic thickness, sufficient to generate anti-magnetic response in a magnetized layer. A plurality of regions on a top surface are activated with a magnetic field in a first direction for a 1 value and in an opposite direction for a 0 value, the multiplication result presented as magnetic field direction on a plurality of output ferromagnetic regions.

Abstract: A superscalar processor has a thread mode of operation for supporting multiple instruction execution threads which are full data path wide instructions, and a micro-thread mode of operation where each thread supports two micro-threads which independently execute instructions. An executed instruction sets a micro-thread mode and an executed instruction sets the thread mode.

Abstract: Embodiments are provided for reduction of lost cycles after branch misprediction in multi-thread microprocessors. In some embodiments, a method includes fetching, by first stage circuitry of a multi-thread microprocessor, a pair of consecutive instructions of a program executed in a thread. The method also includes determining, by second stage circuitry of said microprocessor, during a clock cycle, that a first instruction in the pair is a branch instruction. The method further includes fetching, by the first stage circuitry, during a second clock cycle, a pair of branch target instructions of the program using a branch prediction, and determining, by third stage circuitry of said microprocessor, during the second clock cycle, that the branch prediction is a misprediction. The method still includes sending the second instruction to the second stage circuitry during a third clock cycle, and decoding the second instruction by the second stage circuitry during the third clock cycle.